The baculum in the Vespertilioninae (Chiroptera: Vespertilionidae) with a systematic review, a synopsis of Pipistrellus and Eptesicus, and the descriptions of a new genus and subgenus

(NATURAL HISTORY) ,-;c? The baculum in the Vespertilioninae (Chiroptera: Vespertilionidae) with a systematic review, a synopsis of Pipistrellus and Eptesicus, and the descriptions of a new genus and subgenus J. E. Hill Department of Zoology, British Museum (Natural History), Cromwell Road, London SW7 5BD D. L. Harrison Harrison Zoological Museum, Bowerwood House, St Botolph's Road, Sevenoaks, Kent TN133AL Contents Synopsis Introduction Functional and systematic significance of the baculum Nature and scope of this study Materials and methods Authorship and responsibility The baculum of Pipistrellus The baculum of Eptesicus The baculum in other Vespertilioninae Systematic considerations ..... Genus Pipistrellus Kaup, 1829 Genus Eptesicus Rafi nesque, 1 820 The status of the 'Nycticeini'. The classification of the Vespertilioninae Zoogeographical considerations Conclusions Addendum References Appendix 1 . Specimens examined Tables 1-3 Figures 1-22 Synopsis Current classification of the Vespertilioninae rests chiefly on a suite of mainly adaptive characters, among which facial shortening throughout the subfamily with consequent changes in the structure, size, relative position and number of the incisive and premolar teeth features prominently. Such characters may not necessarily reflect relationships or phyletic diversity, and sometimes do not serve properly to distinguish the genera that they purport to define, as in the distinction of Pipistrellus and Eptesicus, where generic boundaries remain unclear. The search for possibly less strongly adaptive features suggested the possibility that the morphology of the os penis or baculum might prove valuable in the study of the systematics of these genera and perhaps in the subfamily as a whole. This paper reviews earlier studies of the baculum in the Chiroptera and their relevance to systematics in the Order, with an examination of its gross morphology throughout the Vespertilioninae, especial attention being given to species currently allocated either to Pipistrellus or to Eptesicus. A synoptic review of the species Bull. Br. Mus. nat. Hist. (Zool.) 52 (7): 225-305 Issued 30 July 1987 225 226 J. E. HILL & D. L. HARRISON content of these genera is presented, with the recognition and definition of subgenera and included species groups: three such (pumilus, capensis and tenuipinnis) currently referred to Eptesicus on dental grounds seem instead more closely related to Pipistrellus to which they are here transferred. One subgenus of Pipistrellus is described as new (p. 250). The Vespertilioninae as a whole display a wide range of bacular variation, which falls into two major and several minor groups. This has suggested a revision of the current classification of the subfamily, combining bacular features with those conventionally in use. Bacular morphology provides a clear indication that the 'Nycticeini' (or 'Nycticeiini') is an artificial grouping and that the genus Nycticeius as presently understood is composite. Currently it is held to include two species, the North American humeralis and the African schlieffenii: these are here thought to be sufficiently characterised to justify generic separation and a new generic name is proposed for schlieffenii (p. 254). A suggested classification of the subfamily is presented, with a tabulated review of earlier classifications; possible relationships between the constituent genera are discussed and the zoogeography of the bacular types within the subfamily is examined. Introduction A penial bone is known to occur among mammalian Orders in the Insectivora, Chiroptera, Primates, Rodentia and Carnivora. Variously called the os penis, os priapi or os glandis, it was first named the baculum by Thomas (191 5a), the corresponding structure in the female, the os clitoridis, being later called the baubellum by Shortridge (1934: 327, footnote). The features of the baculum have been used extensively in attempts to determine phyletic relationships at various systematic levels (Patterson & Thaeler, 1982). Thomas (loc. cit.), for example, suggested that the baculum might provide evidence valuable in the subfamilial classification of the Sciuridae and indeed pointed out that in this connection there were no bacular features to support the association of the dwarf squirrels in a separate subfamily, the Nannosciurinae. More commonly, bacular features have been used to indicate or determine relationships within genera in the Sciuridae, among New World rodents, and in the Mustelidae. Such characteristics have been employed in species descriptions, especially where bacular variation is pronounced, and also for age determination. Numerous examples of these uses of the baculum are summarised by Patterson & Thaeler (loc. cit.) while Burt (1960) gave an account of the earlier of such studies. Similar early accounts of the baculum in the Chiroptera are reviewed by Hamilton (1949). The presence of a baculum in some at least of the Chiroptera has been long established. Daubenton (1760) described and illustrated (in part) the baculum of Nyctalus noctula and Blainville (1840) similarly studied the baculum of Rhinolophus fermmequinum, R. hipposideros, Vespertilio murinus and again of Nyctalus noctula, the latter author providing perhaps the first accurate and quite detailed drawings of this structure. Later workers such as Ercolani (1868), Robin (1881), Gilbert ( 1 892), Rauther ( 1 903), Gerhardt ( 1 905) and Chaine ( 1 926) provided further details of penial and bacular morphology in the Chiroptera, Chaine in particular discussing and illustrating the baculum in several species and to some extent summarising earlier work in the field. However, none attempted to use the structure of the baculum for systematic purposes. The first use of the baculum in chiropteran systematics appears to be by Thomas (\9\5b) who employed bacular characteristics in defining the species of Nyctophilus. This worker clearly foresaw the value of bacular features in the definition of some at least of the species of bats, beginning from that time a collection of vespertilionid bacula at the British Museum (Natural History) although subsequently making little use of the material that was accumulating, except in 1928 employing bacular characters to separate Indo-Chinese species of Pipistrellus (Thomas, 19280, K). Since then the baculum has been utilised in a variety of taxonomic studies of bats, for example by Krutzsch (1959, 1962) and Lanza (1969) to examine its value in indicating relationships in the Megachiroptera, by Topal (19700) in determining the affinities of la, or by Heller & Volleth (1984) as an indicator of relationship among the species of Pipistrellus and Eptesicus. The baculum of Plecotus was found valuable by Lanza (1960) in discriminating between P. auritus and P. wardi ( = P. austriacus): the subsequent use of the baculum in distinguishing these species is summarised by Corbet (1964). Genoways & Jones (1969) found that bacular features distinguished closely VESPERTILIONINE SYSTEMATICS 227 related species of North American Myotis, LaVal (1973#) employing bacular characters for the same purpose among the Neotropical species of this genus. The emphasis placed on bacular characters in chiropteran systematics is perhaps best illustrated by the number of studies devoted chiefly to bacular structure, often on a regional or faunal basis, as for instance the work by Hamilton (1949) and Krutzsch & Vaughan (1955) on North American species, by Brown et al. (1971) on Neotropical bats, by Topal (1958) on central European species, by Bhatnagar (1967), Agrawal & Sinha (1973), Sinha (1976) and Khajuria (1979, 1980, 1982) on Indian bats, or by Wassif & Madkour (1972) and Wassif, Madkour & Soliman (1984) on Egyptian bats. Bacula are sometimes studied in discrete taxonomic groupings, as for example those of New World molossids by Brown (1967) or of Malaysian Hipposideros by Zubaid & Davison (in press). Thus among the Chiroptera the baculum has been employed as a source of taxonomic features at several systematic levels, but primarily to indicate degrees of relationship or for separation at the specific and sometimes the generic grades, or especially for distinguishing closely related, often sympatric species whose conventional morphological characters are otherwise very similar, as in Myotis and Plecotus. Functional and systematic significance of the baculum Conflicting hypotheses for bacular variation were reviewed in detail by Patterson & Thaeler (1982). These authors proposed that among rodents at least the probability was that the baculum has a precise reproductive purpose and functions primarily as a device contributing to species isolation. Bacular differences among closely related taxa might well then take an exaggerated form. As such, the baculum would be therefore a poor basis for supra-specific classification, but an excellent structure for species diagnosis. Thus they would not consider a phyletic basis for bacular variation to be appropriate. They admit, however, that while in some rodent groups there are patterns of bacular morphology that do not agree with phyletic divergence as indicated by other morphological features, there exist also gross patterns of bacular variation in other groups that do in fact conform with accepted phyletic relationships. Indeed, they remarked that there can be little doubt that the baculum exhibits phyletic weight and consequently may serve as a valuable taxonomic tool. Moreover, taxa that differ in external and cranial characters may have similar bacula, while others that are similar in such features may exhibit highly distinctive genitalia. Patterson & Thaeler (loc. cit.) suggested that although bacular morphology reflects phyletic history on a gross scale, discordance between patterns of bacular and phyletic divergence supports a functional interpretation of bacular variation, especially at the species level. Similarly, opinions vary as to the value of bacular morphology in chiropteran systematics. Hamilton ( 1 949) examined the baculum in North American vespertilionids and concluded that in this family the baculum was useful in defining relationships when considered with skull and other skeletal characteristics. Thus he was able to suggest that the close similarity between the bacula of Myotis (Fig. 19i, j) and Pizonyx (Fig. 19k) indicated their close relationship, and that the dissimilarity between the bacula of Pipistrellus subflavus (Fig. 2d) and P. Hesperus (Fig. 8d) suggested generic or at least subgeneric difference. This author also noted that in most instances among North American vespertilionids there were marked generic differences in the baculum. He considered that further study was needed to determine the usefulness of the baculum in chiropteran systematics and that with time and sufficient material the bone might be utilised in classification. These conclusions were reinforced by Krutzsch & Vaughan (1955) who examined the bacula of further North American species. They remarked that in the case of those that are closely related the baculum can serve as a criterion in judging relationship when other clear cut distinguishing characters are lacking. These authors found bacular variation in closely related bat species to be chiefly in shape, detail of outline, and gross size: their study led to the belief that in at least some superficially similar species well marked and consistent bacular differences reinforced the more subtle external and cranial dissimilarities. Krutzsch (1959) accepted the view that the baculum can provide additional evidence for classification, or, in the absence of other clearly defined characters, can serve as a criterion in judging 228 J. E. HILL & D. L. HARRISON relationship. He added in relation to the Pteropodidae that by virtue of its relative simplicity and structural stability the baculum might well serve to help place entities of doubtful relationship in their natural position, although they might be otherwise morphologically contradictory. Among pteropodids he found that infrageneric differences in the baculum involved minor details of shape, outline and size. Genera, however, might be separated by more profound differences. His study suggested that although within the genus well marked and consistent differences existed between the bacula of individual species, there was nevertheless a basic similarity in pattern throughout the genus, leading to the suggestion that marked variants from this morphological standard in a single genus might provide grounds for a reappraisal of the affinities of the variant. A further study (Krutzsch, 1962) confirmed these opinions, especially in the broad agreement of bacular variation in pteropodids with the taxonomic arrangement of this family by Andersen (1912) and by Tate (1942ft). Krutzsch concluded that strong intrageneric similarities exist among the bacula of pteropodids, but that representative bacula of different genera differ distinctly: although serving well as a source of diagnostic features for the genus, the baculum does not seem to offer exceptional insight into suprageneric relationships. The large genus Pteropus, however, to some extent proves to be an exception, with various of its many species demonstrating considerable variation in bacular structure: on occasion differences between species equal those between some megachiropteran genera. Lanza ( 1 969) examined the baculum of Pteropus in detail and found that its bacular morphology did not conform to the classification proposed by Andersen (1912), a conclusion also reached by Davis ( 1 947) who examined only five species. Lanza found that in many cases bacula of an identical size and shape could be found among species belonging to the same group as well as to different groups; or that the baculum could be extremely different among forms apparently otherwise very closely related. Thus in this genus he found the baculum to be of limited value in phyletic analysis. Similarly, LaVal (1973ft) found that with one exception the bacula of the various species of the vespertilionid Rhogeessa are not sharply differentiated from each other: although in shape they show substantial geographic and individual variation within species they seem nevertheless to differ between species in areas of sympatry or near sympatry. Harrison & Brownlow (1978) found that individual variation in the baculum of adults of another vespertilionid, Scotophilus, was such that it rendered this structure of little or no value in species diagnosis in this difficult genus. Martin (1978) discussed the adaptive value of the baculum in bats, having found a wider range of structural variation among several pteropodid species than was previously thought. He considered that the baculum may have a number of roles of varying adaptive significance in supporting the penis, as a stimulatory structure, or in preventing urethral closure during the pressures of copulation. Although these might allow the baculum to maintain morphological stability within certain taxonomic units, this possible variability of function he thought tended to reduce its value in classification at the specific and subspecific levels. Despite these possible limitations, many authors admit at least the species-specificity of bacular variation among bats, using the baculum to provide additional characters to separate species that sometimes otherwise closely resemble each other. Some examples have been mentioned: others include Wallin (1969) who drew attention to bacular differences in Japanese Pipistrellus and who used such differences to define two species groups in Vesper tilio, or Baag0e (1973) who utilised bacular characters in comparing sibling species of European Myotis. Zubaid & Davison (in press) found the baculum to be specifically diagnostic among Malaysian Hipposideros. In some genera authors have routinely described and illustrated the baculum of new species: for instance Sinha (1969) in describing Pipistrellus peguensis compared its baculum with the bacula of the related species. Similarly, McKean et al. (1978) described and illustrated the baculum of 'Eptesicus' sagittula, comparing it with the bacula of other Australian 'Eptesicus', while Kitchener (1976) employed the baculum of ' Eptesicus' douglasorum in the same way. Bacular characters sometimes form an essential part of revisionary study, as by Kitchener et al. (1986) in defining and keying the Australo-Papuan representatives of Pipistrellus and Falsistrellus. The baculum has also featured in generic revision, Hill (1966a) for example describing and illustrating that of Philetor in the course of such a study, or (1976) that of the majority of the species of Hesperoptenus. Bacular variation has also been employed for generic and subgeneric distinction within the VESPERTILIONINE SYSTEMATICS 229 Vespertilioninae. Wallin (1969) used penial characters in establishing Vespertilio as a genus distinct from Eptesicus and in recognising Hypsugo as a subgenus within Pipistrellus, while Topal (1970a) noted that bacular morphology allied the aberrant genus la more closely to Eptesicus than to Pipistrellus with which it had been associated by some authors. Heller & Volleth (1984) summarised published illustrations of the bacula of Pipistrellus, Eptesicus and some of their associated genera, drawing attention to their taxonomic implications. At a further systematic level, Pine et al. (1971) discussed the penial and bacular morphology of Antrozous and Bauerus in relation to the presumed affinities of these North American genera to the Australian and New Guinea genera Nyctophilus and Pharotis with which they have been associated in the subfamily Nyctophilinae. It is clear from the foregoing account that the baculum is regarded as a valuable source of diagnostic information by many students of chiropteran systematics. This seems especially true in the Vespertilioninae, a subfamily in which diagnosis and definition at both specific and generic levels is sometimes difficult if only the orthodox morphological characters of external, cranial and dental structure are to be relied upon. Nature and scope of this study The basis of the current classification of the Vespertilioninae was first set out in detail by Miller (1907), who recognised a total of thirty-two genera in the group, with diagnoses and short descriptive accounts. The classification of Miller was based chiefly on external, cranial and dental features. Tate (1942a) reviewed the characters used for diagnosis in some detail, dividing the subfamily into four main (tribal) aggregations, and attempting to quantify the interrelationships of its many genera. The major outlines of his classification have since been followed, sometimes with local modification as for instance by Koopman (19840, b, 1985) who subsumed the subfamily Nyctophilinae into the Vespertilioninae. Hill (1966) pointed out that the subfamily comprises a complex of closely interrelated genera separated in some instances by comparatively slender or even rather arbitrary distinctions, the patterns of relationship often obscured by parallelism or convergence. The narrowness of the orthodox distinctions that define many of the constituent genera of the Vespertilioninae has led to much taxonomic combination and recombination since Tate wrote. This situation is exemplified by the more extreme variants of classification that have been proposed. For example, Kuzyakin( 1944, 1950, 1 965) included Pipistrellus and Eptesicus in Vespertilio while Simpson (1945) included Glischropus, Scotozous, Nyctalus and la in Pipistrellus and Rhinopterus, Hesperoptenus, Tylonycteris, Mimetillus, Philetor, Histiotus and Laephotis in Eptesicus. A yet more extreme viewpoint was adopted by Sokolov (1973) who considered that Vespertilio should include not only Pipistrellus and Eptesicus as was thought by Kuzyakin, but also all of the other above mentioned genera except Nyctalus. Horacek & Hanak (19850, b) commented that the concepts of Kuzyakin and Sokolov (with the inclusion of Nyctalus} might be provisionally accepted, at least until factual proof of paraphylly in the group was forthcoming. Nevertheless, they considered this to be a retrograde solution since it expresses nothing of the factual diversity of the group, proposing instead that the problematic taxa should be arranged in separate genera, their diagnoses then making their content clearer though narrower. Both Simpson and Sokolov also included Scotoecus and Scotomanes in Nycticeius as then understood, Baeodon in Rhogeessa, Glauconycteris in Chalinolobus and Dasypterus in Lasiurus to produce a heavily 'lumped' classification. The status of some such as Scotoecus, Dasypterus and Idionycteris has varied from one author to another for decades: in the Australian region Scoteanax and Scotorepens have recently achieved generic rank after many years as nominal subgenera (Kitchener & Caputi, 1985) while la has once again reverted to Pipistrellus (Koopman 1984a, b, 1985) after a brief spell with Eptesicus. The major variants of vespertilionine classification are summarised in Table 1. Many of the characters used to define taxa and relationships among the Vespertilioninae appear strongly adaptive and of equivocal value in generic and suprageneric systematics. Most concern 230 J. E. HILL & D. L. HARRISON ear size and shape, tragal structure, the architecture of the skull, and the number and formation of the teeth. Zima & Horacek (1985) pointed out that the use of the morphological characters employed hitherto in the classification of the Vespertilionidae as a whole might not lead invariably to correct taxonomic conclusions, their degree of differentiation perhaps reflecting the orientation and intensity of selection pressure rather than actual phyletic relationships. These authors indicated an urgent need for new, sufficiently reliable and taxonomically useful criteria based on features that did not possess a direct adaptive significance, including among these the morphology of the reproductive organs and the baculum. Much weight has been placed in the past upon the progressive shortening of the muzzle apparent throughout the Vespertilioninae with concomitant reduction and loss of the incisors and premolars (Tate, 1942a). In the incisive dentition the first upper tooth (i 1 ) 3 is absent, as in all bats. Reduction results in the remaining inner tooth (i 2 ) becoming peg-like and unicuspid, although sometimes quite massive, in a reduction in size of the outer tooth (i 3 ), its displacement inwards or outwards, or in its eventual obsolescence or loss. In the mandible, the first (ij and second (i 2 ) incisor teeth are invariably present, but exceptionally the third (i 3 ) may be absent. The process also involves the reduction and loss of the second upper and lower premolars (pm}) and then of the anterior upper premolar (pm 2 ): thus the premolar formula ranges from pm rf y to pm 5 i 5 f the first upper and lower premolars (pm}) being presumed to be those that are absent from all bats. Seven different combinations of incisors and premolars occur in the subfamily, if Antrozous and Bauerus are included. The full complement is usually taken as the primitive condition, the reduction and disappearance of teeth as derived. These are summarised in Table 2, which gives the incisive and premolar formulae usually attributed to each of the various genera. However, Myotis occur in which pm 3 or pmj are absent (Hill & Topal, 1973), thus in the first instance producing the formula typified by Lasionycteris or Plecotus, in the second the formula for Pipistrellus or Nyctalus; pm 2 may be absent from Pipistrellus to give the arrangement for Eptesicus, or may be present in Eptesicus to produce the formula for Pipistrellus (Hill & Topal, loc. cit.); i 3 is variable in Scotozous (of 45 examined, present in 34, absent from one side or the other in 8, completely absent from 3), when totally absent to produce the incisive formula that usually characterises Nycticeius and its associates (but Thomas & Wroughton ( 1 908) report an example of 'Nycticeius ' schlieffenii in which the left i 3 is present); pm 2 is variable in Scotoecus (Hill, 1974) and in Chalinolobus (Ryan, 1966; Koopman, 1971), and very rarely may be present in 'Nycticeius' schlieffenii (Dobson, 1878; Thomas, 1890). Most genera of Vespertilioninae can be defined by other features besides those of the incisive and premolar dentition, although sometimes only in differing combinations. Thus although some species exist that combine the external features of Myotis with the dental formula of Pipistrellus to the extent that initially they (annectans, ridleyi) were described in the latter genus, other characters such as the form and structure of the tragus and the structure of the incisors enable them to be referred confidently to Myotis (Topal, 19706; Hill & Topal, 1973). Another (rosseti) was first described in Glischropus, subsequently removed to Pipistrellus by Hill (1969) and finally (with ridleyi) to Myotis by Hill & Topal (loc. cit.). However, the genera Pipistrellus and Eptesicus do not offer further conventional characters in this way and are separated for the most part by the presence of pm 2 in the former and its absence in the latter. Wallin (1969) and Hill & Topal (loc. cit.) discussed the variability of this tooth in Pipistrellus and Eptesicus in detail, the latter authors concluding that the presence or absence of pm 2 can have no universal validity in defining the two genera. Heller & Volleth (1984) also examined the relevance of pm 2 in separating Pipistrellus and Eptesicus and concluded that it does not seem to be a reliable characteristic, a classification based on it perhaps misrepresenting true relationship. Tate (\942a) recognised this difficulty but adhered to the conventional practice of separating the two genera by this feature, and indeed the majority of authors have retained the distinction as a matter of convenience, often using the extent of reduction and degree of displacement of pmf from the line of the toothrow as a diagnostic feature between the species of Pipistrellus. "Dental notation of Miller (1907) VESPERTILIONINE SYSTEMATICS 23 1 Koopman (1975) has commented upon this problem. This author examined African species allocated variously to Pipistrellus and Eptesicus in an attempt to find some other character that would divide this large assemblage into two major groups. He could find none among the usual suite of morphological features. Although he found no African Pipistrellus species that closely resembled the hottentotus, tenuipinnis orfloweri groups of Eptesicus and no African member of the latter genus that was similar to thepipistrellus, Hesperus, savii or rueppellii groups of Pipistrellus as he defined them, he did find a resemblance amounting in some cases to virtual identity (if pm 2 was ignored) between the kuhlii group of Pipistrellus and the Eptesicus capensis group. Expanding a view first expressed by Tate (19420) in his account of Eptesicus, Koopman commented that it is probable that the anterior upper premolar has been lost more than once here, and that true phyletic relationships run across the 'generic' line. He thought that it is even possible that in some cases a 'Pipistrellus species' and an 'Eptesicus species' are actually conspecific, but was of the opinion that the available material was insufficient to establish this with certainty for any such pair at the present time. Although retaining Pipistrellus and Eptesicus as separate genera since he believed that the problem should be attacked on a cosmopolitan basis, he remarked that such an arrangement is almost certainly wrong. More recently, Horacek & Hanak (1985-1986) have offered further definitions of Pipistrellus, Hypsugo and Eptesicus. Many varying interpretations can be placed upon external, cranial and dental morphology or on karyological data in the Vespertilioninae. These range from the relationship of one species to another to suprageneric relationships, even to the view that Pipistrellus and Eptesicus may be polyphyletic. Menu (1984), for example, remarked that an exhaustive odontological study of the Vespertilioninae indicated that Pipistrellus includes species wrongly associated by reason of their identical dental formula, but which are not related. Williams & Mares (1978) discussed the karyology of Eptesicus, which as currently defined they thought seemed to be a composite taxon, encompassing perhaps several phyletic lines of pipistrelloid species with reduced numbers of premolars. Heller & Volleth (1984) suggested that Pipistrellus may be a very heterogeneous assemblage and after reviewing the relevance of pm 2 in separating this genus from Eptesicus considered the baculum to be a more reliable guide to the phylogeny of the species of Pipistrellus and Eptesicus, using its features to supplement their findings from karyological data. Many years before this Tate (19420) remarked that it seemed probable that study of the baculum in the Microchiroptera would yield valuable results, with the implication that this might have significance in the classification of the Vespertilionidae. Indeed, Tate records that G. M. Allen had gathered together a number of bacula representing many of the species of Pipistrellus which he intended to employ in revising the genus. Moreover, relatively few species of the nominal genera Pipistrellus and Eptesicus have so far been studied: the impression gained from the literature is that Pipistrellus as currently understood is dignified chiefly by a long, slender shafted baculum and most Eptesicus as it is presently classified by a small, triangular structure, which we have found not to be the case. Initially our intention was to examine the bacula of as many species of Pipistrellus as possible to establish the pattern of bacular variation within the genus, and to compare it with the species groupings proposed by Tate (19420) and by Koopman (1973, 1975). As the work progressed, however, it became increasingly apparent that its implications extended far beyond the limits of this nominal genus and that it was necessary in addition to study the bacula of Eptesicus so far as we were able, and to examine the boundary between these two conventional groupings. Finally, to place our findings in proper perspective, we have surveyed the bacula of most of the remaining genera of the Vespertilioninae and have attempted to assess the generic significance of bacular variation in the subfamily. We have also examined the bacula of Nyctophilus and Pharotis, both usually referred to the closely related subfamily Nyctophilinae. This has been united recently with the Vespertilioninae (Koopman, 19840, b, 1985) and is thus relevant to our study. We have made no detailed examination of the gross morphology of the chiropteran penis except insofar as it is reflected by bacular structures. Nor have we attempted to study its histomorphology. These features are discussed by Smith & Madkour (1980) in an effort to elucidate their relevance to interordinal and infraordinal phylogenetic relationships, and who review earlier studies of penial morphology. 232 J. E. HILL & D. L. HARRISON Materials and methods We have been able to examine bacula from the majority of species currently listed in Pipistrellus and from most of those presently assigned to Eptesicus. In a few instances we have relied upon illustrations and descriptions from the literature. Similarly, for the other genera of the Vespertilioninae our study material has been drawn chiefly from specimens and to a much lesser extent from the published works of others. The specimens that we have examined are listed in Appendix 1 . Our aim as far as genera other than Pipistrellus and Eptesicus are concerned has been to provide illustrations of representative bacula, but in those instances where bacular structure has not before been studied we have endeavoured to examine as many species within each genus as the available specimens permitted. Clearly, the material available to us has been quite inadequate to establish the extent of individual variation in any one species or subspecies. While only adult specimens (wing epiphyses fully fused) have been used, we have necessarily had to accept that for the majority of species our data is limited. We have concentrated therefore on studying and comparing the gross morphology (size, shape, gross structure) of the bacula that we have examined in an attempt to identify similarities, differences and general trends. The finer details perhaps more valuable in species distinction have received much less attention, although where it is known that species are difficult to separate by conventional means attention has been drawn to bacular features that may assist in identification. The specimens used in this study have been drawn almost entirely from the collections of the British Museum (Natural History), London (BM(NH)) and the Harrison Zoological Museum, Sevenoaks, Kent (HZM). Apart from these we have been able to examine one from the Naturhistorisches Museum, Wien (NMW), by courtesy of Dr K. Bauer, and one from the Carnegie Museum of Natural History, Pittsburgh (CMNH), an anomalous specimen loaned for identification by Dr D. A. Schlitter, while Dr K. F. Koopman generously brought to London an example of Nycticeius humeralis from the American Museum of Natural History, New York (AMNH) from which a much needed baculum was obtained. Specimens prepared many years ago at the British Museum (Natural History) are dry, sometimes mounted on card: the remainder have been prepared in the course of this study. This has been accomplished by maceration for a short period in a 5% solution of potassium hydroxide to which a small quantity of alizarin red has been added, after which the grosser macerated tissue was removed by dissection, the specimen then being cleared and stored in glycerin. Drawings have been prepared using either a stereoscopic microscope with graticule scale and attached camera lucida, or freehand using a similar instrument. A few were drawn freehand using a stereo projection microscope with travelling micrometer stage. The wide range of size variation among vespertilionine bacula (for example from a length of 1 mm or less to as much as 9 or 10 mm in Pipistrellus) has necessitated the use of several scales of magnification. So far as possible all drawings on any one page of figures are at the same magnification, with an appropriate scale: to facilitate comparison the varying magnifications used follow an arithmetic progression whereby each successive larger value is twice its predecessor. It has not always been possible to conform to this arrangement, especially where drawings have been prepared from published illustrations. As a rule dorsal (D) and right lateral (RL) views of each baculum are provided: rarely through damage the left lateral (LL) aspect is given. Occasionally where it is of especial interest a half ventral (RVL or LVL) drawing has been made, and in a few instances where drawings have been taken from the literature it has been necessary to give the ventral (V) rather than the dorsal aspect. Authorship and responsibility We take joint responsibility for the results and opinions put forward and expressed in this paper, and for the new names proposed therein. The baculum of Pipistrellus Four bacular types have been identified within the nominal genus Pipistrellus. With some exceptions, modification and combinations, these are in broad agreement with the groupings of species VESPERTILIONINE SYSTEMATICS 233 proposed by Tate (19420) and Koopman (1973, 1975). The classifications of Tate and Koopman are summarised in Table 3. (1) An elongate structure (Fig. la) with a slender shaft and paired basal flanges (e.g. Figs 2a-c, 3, 4, 5), the ventral surface of the proximal part of the shaft transversely concave, its distal part cylindrical or nearly so; in profile the base in line with the shaft or more or less deflected downward at an angle to it; the shaft may be more or less straight, flexed or variously curved in the vertical plane, while the tip is generally bifid or forked and may be directed ventrally to a greater or lesser extent. Species aggregations in which this type of baculum is found include the abramus, pipistrellus, coromandra and tennis groups of Tate (19420); Koopman (1973) amalgamated these to form a pipistrellus group to which he added (1975) the African nanus and permixtus. However, nanus (Fig. 6b) proves to have a very different baculum, as does imbricatus (Fig. 9a), included by Tate in the coromandra group and thus by Koopman ( 1 973) in the pipistrellus group. Pipistrellus babu (Fig. 4a), provisionally placed by Tate in the kuhlii group, also has the long, relatively straight baculum characteristic of this part of the division, as do endoi (Imaizumi, 1959) (Fig. 3b) and peguensis (Sinha, 1969) (Fig. 15c), both described since Tate wrote. The more recently described westralis (Koopman, 1984c) (Fig. lOd), adamsi (Fig. lOc) and wattsi (Fig. lOg) (Kitchener et al, 1986) also belong with tenuis (Fig. 9d) and its allies in this grouping. Taxa referred to the ceylonicus group by both Tate and Koopman (1973) prove to have this bacular structure, as do those that have been examined of the rueppellii group (Figs 7e, f, lOa, b) of Koopman (1975). Pipistrellus kuhlii and its associates (Figs 5a-d, 6c) also belong in this division. In these, however, the basal lobes of the baculum are sharply angled to the shaft in the vertical plane, and this bacular profile is very characteristic of kuhlii and its relatives. The shaft is straight, without flexion, and the tip is usually bifid and not directed ventrally. Koopman (1975) included anchietae (Fig. 6e) in the kuhlii group, but this proves to have a very different bacular configuration. A long-shafted baculum of this type occurs with little modification in the majority of the Australian species (Figs lla-f, 12k) currently referred to Eptesicus, in Nyc talus (Fig. lOf), in Scotozous (Fig. 16d) (to which rueppellii and its immediate associates have sometimes been referred) and in Scotoecus (Fig. 20a-e), in which the 'horns' of the bifid tip extend in some instances almost to form a ring, a condition foreshadowed in Pipistrellus pater culus (Fig. 3c). The Australian Scoteanax (Fig. 16i) and Scotorepens (Figs 16g, h, 21e, also share this bacular type: in Scoteanax the 'horns' at the tip have become a transverse bar, but the species of Scotorepens retain the bifid or slightly bifid tip. A similar long-shafted baculum but with a simple tip occurs in the genera Hesperoptenus (Fig. 21a-c, g) and Chalinolobus (Fig. 17a-e). The baculum of Glischropus (Fig. 18a), although very small, is also of this type, with paired basal lobes, a slender shaft, and bifid tip. (2) A very small structure (Fig. Ib), consisting of a broad base with two basal lobes (e.g. Figs 2d, e, 9c, h), supporting a short, very slightly hollowed shaft. This bacular type is found in subflavus (Fig. 2d), circumdatus (Fig. 2e), societatis (Fig. 9c) and the more recently described cuprosus (Hill & Francis, 1984) (Fig. 9h). (3) A relatively short, stout shafted baculum (Fig. Ic), sometimes with expanded base and tip (e.g. Figs 6a, b, 7a, h, 8e, f), the base on occasion divided into paired lobes, sometimes angled vertically to the line of the shaft, which is fluted ventrally rather than mostly cylindrical; tip when expanded having its anterior edge sometimes divided into several irregular serrations and on occasion downwardly directed. Such bacula are found in the savii group of Tate (19420) and Koopman (1973, 1975) but not in maderensis (Fig. 5b) which was put into the savii group by both authors. Its baculum is however quite different and is like that of kuhlii and its associates. Pipistrellus anchietae (Fig. 6e), referred to the kuhlii group by Koopman (1975) also belongs with savii, and the same bacular type occurs in nanus (Fig. 6b), allocated with permixtus to the pipistrellus group by the same author. We have been unable to examine the baculum of permixtus but that of nanus and ofhelios (Fig. 6d) is of the type characteristic of this division, with its basal part quite sharply flexed to the shaft although not especially deep, and with an expanded, downwardly directed distal part. Pipistrellus eisentrauti 240 J. E. HILL & D. L. HARRISON depression; cranial profile almost straight from occiput to nares, slightly depressed over anterior part of orbit; premaxillae not shortened; zygomata slender, lacking any jugal eminence; interdental palate longer than wide; maxillary toothrows parallel for most of their length, anteriorly slightly convergent; short bony post-palate; slight basial pits; i 2 bicuspid, posterior cusp 1/2-3/4 the height of anterior cusp; i 3 about the same in crown area or a little larger than i 2 , about 1/2 or a little more its height, with larger central and smaller lateral accessory cusps, lying postero-externally to that tooth, separated from c 1 by a small diastema; pm 2 large, unreduced, its crown area similar to that or i 2 or a little less, slightly intruded but separating c 1 and pm 4 ; i l _ 3 not much imbricated, i 3 about twice the bulk of i 1 _ 2 ; pm 2 not usually much reduced, about 1/2-3/4 or more the crown area of pm 4 . Included taxa: aladdin, bactrianus, lacteus, nathusii (Fig. 2b); mediterraneus, (?) permixtus, pipistrellus (Fig. 2a). Among African Pipistrellus we have been unable to examine the baculum of permixtus (Aellen, 1957) compared by its describer chiefly with nathusii. Its dentition, with bicuspid i 2 , the posterior cusp 2/3 the height of the anterior cusp, i 3 with lateral accessory cusps, its main cusp equal in height to the posterior cusp of i 2 , large, slightly intruded pm 2 which is about as big as i 3 , and unreduced pm 2 , its crown area about 3/4-4/5 the crown area of pm 4 suggests that it should be referred to the pipistrellus subgroup. Koopman (1975) referred it to the pipistrellus group. (a) (ii) javanicus (abramus) subgroup. Braincase slightly globular, elevated posteriorly; postorbital region wide; supraorbital region distinctly broadened to produce abruptly incurving lateral margins to the anterior part of the postorbital area; rostrum broad, dorsally flattened, with no more than an indication of a median rostral depression; cranial profile almost straight from occiput to nares, slightly flattened over the occiput and a little depressed over the anterior part of the orbits; premaxillae not shortened; zygomata slender but not weak, lacking any jugal eminence; interdental palate only little longer than wide; palate strongly domed with broad anterior emargination; maxillary toothrows more or less parallel, scarcely convergent anteriorly; short bony post-palate; shallow basial pits; i 2 well developed, bicuspid, posterior cusp sometimes small, usually about 3/4 height of anterior cusp; i 3 similar in size to i 2 or slightly larger, about as high as its posterior cusp, with larger central and smaller lateral accessory cusps, lying postero-externally to that tooth, separated from c 1 by a narrow diastema; pm 2 little reduced, equal to or rather less than i 3 in crown area, in recess between c 1 and pm 4 which approach but do not touch; ij_ 3 scarcely imbricated, i 3 as a rule similar in size to i 2 , both a little more massive than i^ pm 2 about 1/2-3/4 the size of pm 4 , very slightly intruded from toothrow. Included taxa: abramus (Fig. 3a), akokomuli, babu (Fig. 4a), bancanus, camortae (Fig. 15d), endoi (Fig. 3b), irretitus, javanicus (Fig. lOe), meyeni, peguensis (Fig. 15c), paterculus (Fig. 3c), pumiloides. Current treatments of Asian Pipistrellus usually include abramus in P. javanicus (tralatitius , Laurie & Hill, 1954) as a valid subspecies. There appear to be few conventional features that clearly separate javanicus from abramus but their bacula differ quite sharply in the high degree of vertical flexion of the shaft evident in the latter. This difference was used by Thomas (1928a) who examined Indo-Chinese Pipistrellus and differentiated abramus from raptor, javanicus (as tralatitius) and coromandra by virtue of the double curvature of its baculum, the others being straight. Van Peenen et al. (1969) recorded coromandra, javanicus and mimus from Vietnam but the baculum that they illustrate for javanicus is clearly that of abramus. This bacular difference suggests thatjavanicus and abramus should be considered specifically distinct even although there seem to be few cranial and dental characters to separate them. The braincase in javanicus is slightly more inflated than in abramus and its rostrum narrower, the palate is usually a little wider in relation to its length and is slightly more excavated and domed, while pm 2 is a little less reduced and less intruded, tending rather more to separate c 1 and pm 4 . Both occur in Vietnam (Thomas, 1928a; specimens listed below). It seems likely that bancanus and camortae, which has an unflexed baculum, are more closely related to javanicus than to abramus. Soota & Chaturvedi (1980) remarked that Thomas (191 5c) had pointed out that the baculum of abramus is doubly curved and that in paterculus it is straight, but they stated further that material of VESPERTILIONINE SYSTEMATICS 241 paterculus in the collections of the Zoological Survey of India revealed that its baculum is doubly curved. However, specimens in the collections of the British Museum (Natural History) referred to paterculus (some the original material seen by Thomas) have relatively straight bacula when compared with the sinuous baculum of abramus. We have found this sinuous baculum to be characteristic of abramus, to which perhaps the specimens seen by Soota & Chaturvedi should be referred. The very elongate baculum of paterculus, with its strongly bifid tip, the 'horns' of which are deflected ventrally and extend to some extent to form a ring (Thomas, 191 5c) is reminiscent of the baculum of Scotoecus. A very long baculum is also found in endoi, but in this species the tip is less strongly bifid and the 'horns' are deflected dorsally. Both, however, are clearly referable to Pipistrellus on cranial and dental characters, Scotoecus being distinguished especially by a massive unicuspid i 2 , the loss of i 3 , a grooved c 1 , and usually by the absence of pm 2 . (a) (iii) coromandra subgroup. Small, with small, rounded braincase, elevated posteriorly and slightly so frontally; postorbital region wide; rostrum short, relatively narrow; no median rostral depression; cranial profile straight or nearly so from occiput to tip of rostrum; premaxillae exceptionally short; zygomata slender, without jugal projection; interdental palate about as long or a little longer than wide; short bony post-palate: no basial depressions; i 2 usually bicuspid, posterior cusp sometimes very small or rarely absent, when present about 1/2 or a little more the height of the anterior cusp; i 3 equal or greater than i 2 in crown area, reaching to tip of its posterior cusp, with larger principal cusp and smaller lateral accessory cusps, lying postero-externally to the inner tooth; pm 2 not much reduced, nearly as great or as great in crown area as i 3 , with well developed, slightly inwardly directed pointed cusp, in recess between c 1 and pm 4 ; ^ _ 3 not much imbricated, i 3 a little larger than i : _ 2 ; pm 2 about 1/2 crown area and height of pm 4 , slightly extruded. Included taxa: adamsi (Fig. lOc), afghanus, angulatus, collinus (Fig. 4b), coromandra (Fig. 7c), glaucillus, mimus (Fig. 7g), murrayi (Fig. 4c), nitidus, papuanus (Fig. 2c), ponceleti (Fig. 4d), portensis, principulus, sewelanus, sturdeei; possibly subulidens which may however represent javanicus; tenuis (Fig. 9d), tramatus (Fig. 7b), wattsi (Fig. lOg), westralis (Fig. lOd). (a) (iv) ceylonicus subgroup. Large, with rather short, broad braincase; wide postorbital region; some degree of supraorbital expansion; rostrum broad, rather long; weak, diffuse median rostral depression; cranial profile slightly convex, raised over the frontal region; premaxillae normal, not shortened; zygomata moderate, without jugal eminence or process, interdental palate longer than wide; maxillary toothrows parallel; short bony post-palate; slight basial pits; i 2 large and massive, bicuspid to almost unicuspid, with moderate to small posterior cusp about 2/3 height of anterior cusp; i 3 massive, as large or larger than i 2 , extending to or a little beyond posterior cusp of that tooth, with large principal cusp and smaller lateral accessory cusps, lying postero-laterally to i 2 , narrowly separated from c 1 ; pm 2 large, nearly as great or greater in crown area than i 3 , usually filling the recess between c 1 and pm 4 into which it is intruded, these almost in contact labially; i : _ 3 slightly imbricated, i 3 a little larger than i 1 - 2 ' pm 2 almost as large in crown area as pm 4 , very slightly extruded from the toothrow. Included taxa: borneanus, ceylonicus (Fig. 7d), chrysothrix, indicus, (?) minahassae, raptor (Fig. 3d), shanorum, subcanus. An account of minahassae is given by Tate (\942a) who referred it to a minahassae group of which it was the sole member. The skull of the holotype has never been described and Tate's remarks are based on a referred specimen in the American Museum of Natural History, New York (AMNH 102359). It has a short, high braincase with rudiments of a sagittal crest, prominent supraorbital tubercles and slender zygomata; i 2 is long, with well developed posterior cusp, c 1 slender, lacking an accessory cusp, pm 2 only slightly intruded, its crown area greater than that of i 3 , and i^-j scarcely imbricated. These features suggest that if this specimen represents minahassae the taxon should be allocated to Pipistrellus (Pipistrellus) and provisionally we place it in the ceylonicus subgroup of the pipistrellus group, but clearly these decisions can only be speculative. (b) rueppellti group Baculum as in pipistrellus group; braincase high, broadened, rounded and globose; postorbital 242 J. E. HILL & D. L. HARRISON region wide; supraorbital region slightly expanded; rostrum short; with shallow, ill-defined median depression; cranial profile almost straight, a little raised over frontal region, a little depressed over rostrum; premaxillae not shortened; zygomata slender, without jugal projection; interdental palate a little longer than wide; maxillary toothrows slightly convergent; short bony post-palate; no basial pits, instead a shallow depression; i 2 strongly bicuspid, posterior cusp about 3/4 height of anterior cusp; i 3 usually very small or minute, its crown area less than 1/2 that of i 2 , its tip sometimes barely rising above the cingulum of the inner tooth, on occasion (e.g. nanulus) larger, equal to or slightly exceeding i 2 in crown area, about 1 /2 or a little more the height of that tooth; i 3 lying sublaterally to i 2 , separated from c 1 by a wide diastema; pm 2 not usually greatly reduced, its crown area similar to that of i 2 , with strong cusp, separating c 1 and pm 4 , occasionally (crassulus) much reduced, similar in size to i 3 in its much reduced condition, or (crassulus, nanulus) recessed between these teeth; i t _ 3 little imbricated, i 3 slightly the largest as a rule; pm 2 about 3/4 or more as large in crown area as pm 4 and about 3/4 its height, rarely (coxi, crassulus) more reduced, about 1/2 crown area and height of pm 4 . Included taxa: Probably coxi', crassulus (Fig. le),fuscipes, leucomelas, nanulus (Fig. lf),pulcher (Fig. lOa), rueppellii (Fig. \0b),senegalensis, vernayi. Vansonia Roberts, 1946 is available should further separation of the rueppellii group be thought justified: an earlier name, Alobus Peters, 1867 is preoccupied. (c) kuhlii group Baculum of moderate length with narrow cylindrical shaft and paired basal lobes as in pipistrellus and rueppellii groups but basal lobes strongly angled to line of shaft in vertical plane; braincase low but not flattened, rounded, only slightly elongate; postorbital region wide; supraorbital region not widened or swollen; rostrum long, unwidened, with very slight median flattening; cranial profile almost straight from occiput to nares, slightly raised over frontal region, slightly depressed over front of orbits; premaxillae slightly shortened; zygomata slender, weak, without jugal eminence; interdental palate longer than wide; maxillary toothrows almost parallel; short bony post-palate; small, narrow basial pits; i 2 usually unicuspid, at best only slightly bicuspid; i 3 small, its crown area 1/2 or less that of i 2 , its tip extending only slightly beyond the cingulum of that tooth, to which it lies laterally or sublaterally, separated from c 1 by a moderate or narrow diastema; pm 2 small, similar in crown area to i 3 , intruded to lie in recess between c 1 and pm 4 , these more or less in contact; i t _ 3 moderately imbricated, i 3 slightly the largest; pm 2 reduced, about 1/2 or less the crown area and height of pm 4 . Included taxa: Probably aero; deserti(aegyptius, Qumsiyeh, 1985) (Fig. 5c),fuscatus, ikwanius; probably inexspectatus; kuhlii (Fig. 5a), maderensis (Fig. 5b), marrensis, rusticus (Figs 5d, 6c). We have been unable to examine the baculum of inexspectatus (Aellen, 1959) but this taxon was placed in the kuhlii group by Koopman (1975) who also referred maderensis to the savii group. However, an example of maderensis in the collections of the British Museum (Natural History) has a baculum clearly of the kuhlii type. Romicia Gray, 1838 is available for the kuhlii group should this be thought worthy of further separation. Subgenus Pipistrellus ( Vespadelus) Baculum usually with long cylindrical or ventrally slightly fluted shaft, paired basal lobes and a blunt tip; shaft shorter and wider in sagittula; basal lobes sometimes flexed to line of shaft in vertical plane; braincase slightly broadened, flattened and elongated; postorbital region wide; supraorbital region slightly broadened; rostrum short but not greatly widened; shallow median rostral depression; cranial profile almost straight from occiput to nares, a little depressed over rostrum; premaxillae not shortened; zygomata slender, without jugal process; interdental palate a little longer than wide; maxillary toothrows slightly convergent anteriorly; short bony post-palate; no basial pits: i 2 bicuspid, posterior cusp almost as high as anterior cusp; i 3 much reduced, its crown area 1 /2 or less that of i 2 , its tip barely extending beyond the cingulum of that tooth, to which i 3 lies postero-laterally, separated from c 1 by a narrow diastema; pm 2 almost invariably absent, VESPERTILIONINE SYSTEMATICS 243 when present a small spicule in recess between c 1 and pm 4 : i { _ 3 moderately imbricated, i 3 slightly the largest, pm 2 greatly reduced, in crown area about 1/2 or more usually less the crown area of pm 4 , and 1/2 its height. Included taxa: caurinus (Fig. lib), douglasorum (Fig. 1 ld),pumilus (Figs 1 la, 12k), regulus (Fig. 1 le), sagittula (Fig. 1 1 0, vulturnus (Fig. 1 Ic). Formerly referred to Eptesicus, the transfer of these taxa to Pipistrellus was first suggested by Heller & Volleth (1984), purely on bacular grounds. The bacular, cranial and dental features of this group suggest that it represents P. (Pipistrellus) in Australia, the few members of this subgenus (adamsi, westralis and perhaps javanicus) that also occur there being possibly slightly less differentiated by virtue of their relatively slightly less shortened skulls and their retention of pm 2 . The pipistrellus group of the subgenus extends widely through the islands of IndoAustralia to New Guinea, the Solomon Islands and Australia, chiefly as the coromandra subgroup, to which adamsi and westralis belong. Thejavanicus subgroup reaches at least to Java and Sulawesi and may extend to Australia (Hill, 1983) but the Australian record ofjavanicus is based on two old examples and has never been confirmed. Possibly the slightly differentiated pumilus and it allies result from a further perhaps earlier invasion of Australia. Bacular differences in this subgenus (Figs 11, 12k) suggest that it may consist of two groups: it has been possible to examine only pumilus. There has been hitherto a wide geographical hiatus in the Indo-Australian distribution of Eptesicus as formerly understood. Beyond these Australian forms, no other taxon attributed to this nominal genus has been reported further east in Indo-Australia than southern Thailand, other than an unconfirmed record from Sarawak of Eptesicus sp. (Pirlot, 1968) which provided no details. Subgenus Pipistrellus (Perimyotis) Baculum very small, Y-shaped, with paired basal lobes and very short shaft; braincase slightly elongate, rounded, almost globose; postorbital region wide; supraorbital region slightly broadened; rostrum long, elevated, slightly widened; shallow median frontal depression; a very slight lateral depression on each side just anterior to the orbital rim; cranial profile sinuous, raised over frontal region, a little depressed over front of orbits; premaxillae not shortened; zygomata moderate, a slight jugal eminence; interdental palate longer than wide with wide anterior palatal emargination; maxillary toothrows convergent anteriorly; very short bony post-palate; slight basial depressions; i 2 bicuspid with well developed posterior cusp about 3/4 height of anterior cusp; i 3 massive, its crown area exceeding that of i 2 , in height reaching or exceeding the height of anterior cusp of that tooth, with larger principal cusp and smaller lateral accessory cusps, lying posteroexternally to inner tooth, separated from c 1 by a wide diastema; pm 2 large, its crown area equal to that of i 3 , in toothrow, sometimes separated from pm 4 by a slight diastema; i 2 " 3 and pm 2 almost identical to those of P. nathusii; i^_ 3 not imbricated, i 3 only slightly bulkier than i : _ 2 ; pm 2 not greatly reduced or compressed in toothrow, its crown area about 1/2 or more that of pm 4 , about 1/2-3/4 its height; tragus myotine, about 1/2 height of ear, tapering to blunt point. Included taxon: subflavus (Fig. 2d). Menu (1984) proposed the genus Perimyotis for P. subflavus, chiefly on account of the features of the canine and post-canine dentition in which he believed this species to approach Myotis. However, Hill & Topal (1973) in discussing Myotis rosseti and M. ridleyi which also combine the tragal features of Myotis with the Pipistrellus dentition (pm| absent) noted that in Myotis i 2 is short and broad, its posterior cusp wider basally than the anterior cusp, while in Pipistrellus this tooth is linear, often narrower posteriorly than anteriorly. Also, in Myotis the principal cusp of i 3 is equal to or exceeds that of i 2 in height and the tooth is often hooked to produce a caniniform apperance while in Pipistrellus it is lower and is not hooked. In Myotis i 3 is usually much larger than i l _ 2 but in Pipistrellus there is as a rule no such great distinction in size. The incisive dentition of subflavus corresponds closely with that of Pipistrellus. The baculum of subflavus is of a type not found in Myotis. Menu (1984) stated on the basis of published figures that the baculum approached that of certain Myotis and more particularly that of Plecotus auritus. We find no significant resemblance to the morphologically rather stable, saddle- 244 J. E. HILL & D. L. HARRISON like baculum ofMyotis and although there are some similarities with the bacula ofPlecotus auritus (Fig. 19g) and P. teneriffae (Fig. 14d), that of P. austriacus (Fig. 19h) is nearer in structure to the myotine baculum. The bacular type found in subflavus occurs in a similar form in Pipistrellus circumdatus (Fig. 2e), P. societatis (Fig. 9c) and P. cuprosus (Fig. 9h). There are considerable differences, however, between subflavus and circumdatus and its allies, not least in the degree of reduction of pm 2 , this tooth in these three species being very small or absent. The unshortened rostrum and the dental features of subflavus suggest that it is nearest to P. (Pipistrellus), which it appears to represent in North America. We find hesperus, the other North American species of Pipistrellus, to belong on bacular and dental grounds to P. (Hypsugo). Thus we do not support Menu's view (p. 410, footnote) that Pipistrellus is limited to the Old World and that the lines leading to this genus did not enter the North American continent. The marked differences between subflavus and hesperus indicate two quite different pipistrelline groups, as Hamilton (1949) remarked in relation to their bacula, but bacular and dental evidence suggests alliance to established Old World groupings, the baculum of subflavus being perhaps a reduced form of the shafted structure found in P. (Pipistrellus) , that of hesperus a modification of the type found in P. (Hypsugo) . Subgenus Pipistrellus (Hypsugo) Baculum usually short, stout, sometimes with expanded base and tip; base rarely bilobed, sometimes dorsally elevated; shaft generally flattened dorso-ventrally, sometimes wide, its underside transversely concave or fluted; tip ventrally hollowed as an extension of ventral fluting of shaft, when expanded anteriorly sub-square or slightly rounded, its anterior edge sometimes irregularly serrated, tip sometimes downwardly directed, its lateral margins on occasion forming two broadly based, ventrally directed projections; pm 2 generally much reduced, small, minute, or rarely absent. Wallin ( 1 969) considered Hypsugo a valid subgenus within Pipistrellus but included only P. savii: Horacek & Hanak (1985#, b) added cadornae and pulveratus and suggested the elevation of Hypsugo to generic rank, subsequently (1985-1986) widening its possible content and considering it generically distinct. (a) savii group Postorbital region, supraorbital region and rostrum not greatly widened; supraorbital tubercles if present small and undeveloped. (a) (i) pulveratus subgroup. Braincase elongate, inflated; postorbital region wide, supraorbital area not broadened; rostrum long, not widened; shallow frontal depression; no median rostral depression; broad, shallow lateral depressions above anterior part of orbit; cranial profile somewhat sinuous, depressed over front of orbits; premaxillae not shortened; zygomata robust, with very slight jugal eminence; interdental palate longer than wide; maxillary toothrows almost parallel; moderate bony post-palate; no basial pits; i 2 bicuspid, posterior cusp about 3/4 height of anterior cusp; i 3 large, wide, its crown area equal to or slightly exceeding that of i 2 , reaching to tip of the posterior cusp of that tooth, with moderate lateral accessory cusps, lying postero-externally to the inner tooth, separated from c 1 by a moderate diastema; pm 2 about equal or nearly equal to i 3 in crown area, in recess between c 1 and pm 4 which are closely approximated; ^.3 slightly imbricated, i 3 a little the largest; pm 2 a little less than 1/2 the crown area of pm 4 , 1/2-3/4 its height. Included taxon: pulveratus (Fig. 8c). (a) (ii) nanus subgroup. Braincase elevated, slightly inflated, more or less globose but a little elongated; postorbital region wide; supraorbital area slightly widened with small supraorbital swellings; rostrum not especially shortened or broadened; shallow median rostral depression; slight lateral depressions just anterior to supraorbital region; cranial profile sinuous, strongly depressed and concave over rostrum; premaxillae not shortened; zygomata slender, lacking jugal projection; interdental palate longer than wide; maxillary toothrows slightly convergent; short bony post-palate; no basial pits; i 2 unicuspid or with small posterior cusp extending to about 3/4 of VESPERTILIONINE SYSTEMATICS 245 its height; i 3 wide, its crown area slightly exceeding that of i 2 , about 1/2-3/4 the height of that tooth, extending almost to the tip of its posterior cusp, with slight lateral cusps, lying postero-externally to the inner tooth, separated from c 1 by a wide diastema; pm 2 about 1/2-2/3 or a little more the crown area of i 3 , intruded into recess between c 1 and pm 4 , these sometimes in contact or nearly so; i : _ 3 not or only very slightly imbricated, i 3 slightly the largest; pm 2 about 1/2 the crown area and height of pm 4 . Included taxa: arabicus (Fig. 7a), culex, Helios (Fig. 6d); probably musciculus; nanus (Fig. 6b), stampflii. Current listings (i.e. Hayman & Hill, 1971; Koopman, 1975) unite Helios with P. nanus as a synonym or possibly as a valid subspecies. However, the bacular features of this pale form suggest that it may represent a species distinct from nanus with which it may be sympatric in northern and eastern Kenya and in the Sudan. No baculum has been available for musciculus, which was placed in a Hesperus group by Koopman (loc. cit.). Although its incisive and premolar dentition agrees with the nanus subgroup its placement here remains speculative. (a) (in) savii subgroup. Braincase rather low and flat, elongate rather than globose; postorbital region not especially widened; supraorbital region unwidened or only slightly widened; rostrum of moderate length; a shallow median rostral depression; usually slight lateral rostral depressions just anterior to supraorbital and anterior orbital rim; cranial profile straight or slightly concave; premaxillae not shortened; zygomata robust, often with slight jugal process; interdental palate a little longer than wide; maxillary toothrows more or less parallel; short bony post-palate; shallow or no basial pits; i 2 unicuspid or with posterior cusp, when present about 3/4 height of anterior cusp; i 3 similar to or exceeding i 2 in crown area, about 1/2-3/4 the height of i 2 , with strong central cusp flanked by smaller lateral accessory cusps, lying postero-externally or more laterally (anchietae) to the inner tooth, separated from c 1 by a strong diastema; pm 2 much reduced, minute or absent, when present crown area less than 1/2 that of i 3 , in recess between c 1 and pm 4 , these as a rule in contact; \^ _ 3 slightly or more strongly imbricated, similar in size or i 3 slightly the bulkiest; pm 2 reduced, about 1/2 or less in crown area than pm 4 and about 2/3 its height. Included taxa: anchietae (Fig. 6e); probably ariel', probably austenianus; bodenheimeri (Fig. 9f), caucasicus, darwini, maurus, savii (Fig. 6a). We have been unable to examine the baculum of ariel. The baculum of a small Pipistrellus from the Naturhistorisches Museum, Wien (from Sayala, Upper Egypt) tentatively identified as ariel is illustrated by Gaisler et al. (1972) but is evidently of the kuhlii type. Qumsiyeh (1985) employs the description of this baculum in his account of ariel. However, Dr K. Bauer informs us (in litt.) that the specimen (NHW 10351) of which the baculum is figured by Gaisler et al. (loc. cit.) is not referable to ariel but is instead a small deserti, an identification clearly supported by its bacular structure. Moreover, Dr Bauer has loaned three similarly small specimens, one male, the others female (NHW 27501-3) (length of forearm 29-2, 28-9, 28-2; condylobasal length 11-0, 10-5. 10-9; c-m 3 4-0, 3-8, 3-9) apparently from Upper Egypt, that also represent deserti: a baculum from this sample is again exactly of the kuhlii type. The cranial (narrow braincase, unexpanded rostrum, short broad narial and anterior palatal emarginations, narrow basioccipital) and dental (long i 3 , minute pm 2 ) features of ariel clearly indicate that it belongs with savii, to which group Koopman (1975) referred it. \syntype of Eptesicusbicolor(Bocage, 1889)(BM(NH) 89.5. 1.3) (Fig. 9e) proves to be identical cranially, dentally and in bacular morphology with Pipistrellus anchietae (Seabra, 1900) (vide infra, p. 249). However, the point needs confirmation or otherwise by examination of the other syntype in the Museu Nacional de Lisboa. It should be noted that bicolor is the prior name (HonackietaL, 1982). The relationship between the pulveratus , nanus and savii subgroups is illustrated by arabicus and bodenheimeri, the bacula of which are compared directly by Harrison (1982). The baculum of arabicus (Fig. 7a) approaches that of anchietae (Fig. 6e) yet cranially and dentally this species is nearer to nanus (Fig. 6b), while that of bodenheimeri (Fig. 9f) is like the baculum of pulveratus (Fig. 8c) but cranially and dentally the species is close to savii (Fig. 6a). These combinations of features link the three subgroups. 246 J. E. HILL & D. L. HARRISON (a) (iv) Hesperus subgroup. Baculum a fluted structure, much like that ofpulveratus or bodenheimeri. Braincase low but broad, elongated; postorbital region wide; supraorbital area slightly widened; rostrum short, not greatly broadened; a shallow median frontal depression; slight lateral rostral depressions just above anteorbital foramina; cranial profile almost straight, slightly depressed above anterior root of zygomata; premaxillae not shortened; zygomata slender, a little widened anteriorly, lacking any jugal eminence; interdental palate about as wide as long; maxillary toothrows convergent; short bony post-palate; no basial pits; cochlear bullae inflated with narrow basioccipital; i 2 unicuspid; i 3 slightly greater in crown than i 2 but about 1/2 its height, with little trace of lateral accessory cusps, lying postero-externally, separated from c 1 by moderate to small diastema; pm 2 small to minute, at best about 1/2 or less in crown area than i 3 , in recess between c 1 and pm 4 which are closely approximated; ij_ 3 scarcely or not imbricated, similar in size; pm 2 reduced, about 1/2 the crown area of pm 4 , a little less than 1/2 its height. Included taxon: Hesperus (Fig. 8d). Horacek & Hanak (1985a, b, 1985-1986) have indicated that they intend to propose generic status for Hesperus and indeed have suggested that it be referred to Parastrellus which they offer as a new name. It is however a nomen nudum in these publications. There seem good grounds for considering Hesperus the North American representative of P. (Hypsugo) to which its bacular, cranial and dental features ally it. Like bodenheimeri (Fig. 9f) its baculum approaches that of pulveratus (Fig. 8c) but cranially and dentally it is nearer to savii&nd its immediate allies. Koopman (1975) referred Hesperus to a Hesperus group in which he also included the African musciculus, here provisionally allocated to the nanus subgroup. (a) (v) eisentrauti subgroup. Braincase broad, elevated and globular; inflated frontally; postorbital region wide; supraorbital region broadened, with small supraorbital tubercles; rostrum short, deep, wide and massive; slight median rostral depression; cranial profile straight or slightly convex; premaxillae not shortened; zygomata strong, lacking any jugal projection; interdental palate very slightly longer than wide; maxillary toothrows almost parallel; short bony post-palate; slight basial pits usually present; i 2 long, narrow, bicuspid, posterior cups about 3/4 height of anterior cusp; i 3 short, wide, similar to or slightly greater in crown area than i 2 , about 1/2 or a little more its height, with larger central cusp and smaller lateral accessory cusps, lying laterally and slightly posteriorly to the inner tooth, separated from c 1 by a moderate diastema; pm 2 small, about the same in crown area as i 3 , sandwiched into recess between c 1 and pm 4 , these almost in contact; i t _ 3 slightly imbricated, i 2 _ 3 similar in size, both larger than i t ; pm 2 about 1/2 crown area and height of pm 4 . Included taxon: eisentrauti (Fig. 9g). Koopman (1975) places eisentrauti in a rueppellii group, no doubt on account of its elevated, inflated braincase and its bicuspid i 2 , but its bacular features do not associate it with this species and its immediate allies. Its baculum is very similar to that of imbricatus (Fig. 9a) or macrotis (Fig. 9b). (a) (vi) imbricatus subgroup. Braincase inflated, globular, raised posteriorly; postorbital region Wide; supraorbital area slightly widened with very small supraorbital tubercles; rostrum short, not especially broadened; no median rostral depression; cranial profile almost straight; slightly concave above supraorbital region; premaxillae not shortened; zygomata moderate to strong, sometimes with a trace of a jugal eminence; interdental palate about as wide as long, not domed; maxillary toothrows almost parallel; very short bony post-palate; well developed basial pits; i 2 bicuspid, posterior cusp about 3/4 height of anterior cusp; i 3 similar in crown area to i 2 , about 1/2 its height, with larger central cusp and smaller lateral accessory cusps, lying laterally to the inner tooth, separated from c 1 by a narrow diastema; pm 2 greatly reduced, 1/4 or less the crown area of i 3 , in recess between c 1 and pm 4 , these in contact; i l _ 3 scarcely imbricated, i 2 _ 3 of similar size, a little larger than i t ; pm 2 about 1/2 the crown area and height of pm 4 . Included taxa: curtatus, imbricatus (Fig. 9a), macrotis (Fig. 9b), vordermanni. (a) (vii) lophurus subgroup. Braincase inflated, rounded, slightly elongate, raised posteriorly; postorbital region wide; supraorbital area little widened; at best only a trace of supraorbital VESPERTILIONINE SYSTEMATICS 247 tubercles; rostrum moderate in length, longer than in imbricatus subgroup, not broadened; no median rostral depression; cranial profile almost straight, slightly depressed or concave above supraorbital region; zygomata strong with distinct jugal eminence; interdental palate a little longer than wide; maxillary toothrows slightly convergent; moderate bony post-palate; deep basial pits; incisor and premolar dentition closely similar to that of imbricatus subgroup but i 3 lying more postero-laterally to i 2 , and pm 2 sometimes (lophurus) slightly larger, about 1/2 crown are of i 3 . Included taxa: cadornae, kitcheneri (Fig. 8e), lophurus (Fig. 80The baculum of kitcheneri is unusual in the presence distally of two anterior dorso-lateral, posteriorly directed processes, with ventrally a more or less tapered median gutter. As in lophurus, the tip is directed slightly ventrally. (b) stenopterus group Braincase large, rounded and globular; postorbital region very wide; supraorbital region much widened to include well developed supraorbital tubercles; rostrum short, wide; shallow median rostral depression anterior to frontal region; cranial profile slightly convex, elevated over frontal area; premaxillae not shortened; zygomata rather weak, lacking jugal process but usually with small descending process external to m 3 ; palate short and broad, the interdental palate as wide as long; maxillary toothrows parallel or nearly so; short bony post-palate; shallow basial pits; i 2 small, bicuspid, posterior cusp 1/2-3/4 height of anterior cusp; i 3 a little smaller in crown area than i 2 , its tip reaching almost to tip of the posterior cusp of that tooth, with large central cusp and smaller lateral accessory cusps, lying postero-laterally to the inner tooth, only narrowly separated from c 1 or almost in contact with it; c 1 with distinct, well defined posterior accessory cusp; pm 2 small or minute, about equal in crown area or a little larger than i 3 (stenopterus) or about 1/3-1/4 the crown area of this tooth (joffrei, anthonyi), in recess between c 1 and pm 4 , which touch; i t _ 3 not much imbricated, all of similar size; crown area of pm 2 slightly exceeding that of pm 4 , pm 2 similar in height to the second tooth (stenopterus), or crown area of pm 2 about 1/2 that of pm 4 , pm 2 almost as high as that tooth (joffrei, anthonyi). Included taxa: anthonyi, joffrei, stenopterus (Fig. 7h). The baculum of stenopterus is unusual, although of the savii type: it has a narrow lobed base, hollowed shaft, and expanded tip the lateral margins of which project ventrally as two broadly based 'horns'. The stenopterus group as here understood is the joffrei group of Tate ( 1 942a) and (in part) of Koopman (1973). Both joffrei and stenopterus have been referred in the past to Nyctalus but as mentioned above (p. 234) the baculum of stenopterus has no resemblance to the long-shafted baculum of that genus (Fig. 100Tate (19420) referred both to Pipistrellus with the comment that the group approached Oriental members of the savii group, and might at a later time be accorded generic rank. Subgenus Pipistrellus (Falsistrellus) Baculum a broad, proximally widened and ventrally deeply fluted structure with no distal expansion; braincase elongate; postorbital region wide; supraorbital area not expanded; rostrum long, not broadened; zygomata moderate to strong; and palate rather narrow, the interdental palate longer than wide. Pipistrellus (Falsistrellus) appears to be related to P. (Hypsugo) of which it may be the eastern representative. It is approached in bacular morphology by some of the latter subgenus such as imbricatus (Fig. 9a), macrotis (Fig. 9b), kitcheneri (Fig. 8e) and lophurus (Fig. 8f), and indeed the baculum in P. (Falsistrellus) appears to be an extreme variant of the broad, ventrally fluted structure of many of P. (Hypsugo). Kitchener et al. (1986) raised Falsistrellus to generic rank but did not include affinis andpetersi, confining their comparisons to the Australasian Pipistrellus (i.e. adamsi, angulatus, collinus, papuanus, wattsi and westralis) here referred to the coromandra subgroup of P. (Pipistrellus). These authors drew attention to its larger size; to its small i 3 which is anteriorly displaced and swivelled or rotated outwards to lie alongside i 2 , its concavity facing outwards (a feature which may have influenced Iredale & Troughton, 1934 in placing it in Glischropus); and to its combination of 248 J. E. HILL & D. L. HARRISON unicuspid i 2 , tiny pm 2 and pronounced occipital crest, which asTate (19420) noted gives the rear of the skull a 'helmeted' appearance. Excepting the large size and the presence of a strong occipital crest, these features occur elsewhere in the various groups of Pipistrellus: the extent of the occipital crest may be a function of the large size of the skull. (a) affinis group Braincase rather narrow, mastoid width markedly less than zygomatic width; postorbital region wide; slight, rather poorly developed supraorbital ridges; very shallow median rostral depression just anterior to frontal region; dorso-lateral margin of rostrum more or less straight from postorbital constriction to anterior edge of orbit; cranial profile almost straight, slightly elevated frontally, depressed over supraorbital region; premaxillae not shortened; zygomata moderate to robust with jugal eminence; maxillary toothrows slightly convergent; moderate bony post-palate; no basial pits; i 2 strongly bicuspid, posterior cusp 3/4 or more the height of the anterior cusp; i 3 larger in crown area than i 2 , its height about equal to the height of the posterior cusp of that tooth, with strong central cusp and smaller lateral accessory cusps, lying postero-laterally to the inner tooth, separated from c 1 by a moderate diastema; pm 2 almost as great in crown area as i 3 (affinis) or about 3/4 its crown area (petersi), in recess between c 1 and pm 4 ; i^ _ 3 moderately imbricated, i 3 the largest; pm 2 about 1/2 the crown area of pm 4 and about 2/3-3/4 its height. Included taxa: affinis (Fig. 8a), (?) mordax, petersi (Fig. 8b). It has not been possible to examine mordax. Indian records of this species appear to be based on specimens in the collections of the British Museum (Natural History) tentatively labelled as such. These, however, agree closely with the description of P. affinis by Dobson (1871) and with the account of a specimen referred to this species from Likiang, Yunnan by Tate (19420). If correctly allocated, mordax (Peters, 1867) is the earliest name in the group. (b) tasmaniensis group Large and distinctive; braincase high, with well developed sagittal crest; postorbital region wide and strong; no median rostral depression; cranial profile straight; premaxillae slightly shortened; zygomata strong with slight jugal process and small inferior process; maxillary toothrows nearly parallel; short bony post-palate; slight basial depressions; i 2 large, unicuspid; i 3 small, its crown area about 1 /4 that of i 2 , barely extending above the cingulum of that tooth to which it lies laterally, its hollowed face outwardly directed, separated from c 1 by a moderate diastema: pm 2 very small, about 1 /3 the crown area of i 3 , in recess between c 1 and pm 4 , which touch; i l _ 3 much imbricated, i 3 twice the bulk of i l - 2 \ pm 2 much reduced, about 1/4 the crown area of pm 4 and about 1/2 its height. Included taxa: mackenziei (Fig. lOh), tasmaniensis (Fig. 8g) Subgenus Pipistrellus (Neoromicia) Baculum with distinct paired basal lobes, slender cylindrical shaft and variously expanded tip; braincase broad, sometimes slightly elongate, rather flattened; postorbital region wide; supraorbital area unwidened or only slightly broadened; rostrum moderate or slightly lengthened; cranial profile straight; premaxillae shortened; zygomata moderate, no jugal process; maxillary toothrows only slightly convergent; short bony post-palate; no basial pits; i 2 unicuspid or with small posterior cusp extending for about 3/4 its height; i 3 smaller than i 2 , its crown area 3/4-1/2 or less that of the inner tooth, about 1/2 its height, with usually a larger central cusp and slight lateral accessory cusps, the inner cusp as a rule very small, the tooth anteriorly displaced to lie alongside or almost alongside i 2 , separated from c 1 by a moderate to wide diastema; pm 2 almost invariably absent, when present very small, in recess between c 1 and pm 4 , i l _ 3 slightly to moderately imbricated, of similar size of with i 2 _ 3 a little the larger; pm 2 reduced, its crown area 1/2 or less that of pm 4 and its height 1/2-3/4 of the height of that tooth. This subgenus is wholly African and hitherto its members have been referred to Eptesicus, although there is karyological evidence (vide infra) suggesting that one at least should be moved to Pipistrellus. It incorporates the Eptesicus capensis and E. tenuipinnis groups of Koopman (1975). VESPERTILIONINE SYSTEMATICS 249 These can be recognised readily by the structure of the baculum, capensis and its allies (Fig. 12a-d, f-i) having the distal part of the baculum spatulate and ventrally deflected, tenuipinnis and its associates (Fig. 12e, j) having the baculum modified distally into a lobed, almost vertical plate-like structure. Published karyological data refers only to capensis, although studies of other members of the subgenus are in progress (Rautenbach & Schlitter, 1985a, b). Peterson & Nagorsen (1975) found that capensis has a diploid number of 32 and a fundamental number of 50: Williams & Mares (1978) discussed the possible composite nature of Eptesicus as suggested by Koopman (1975) and pointed out that the species fitted karyologically within the variation exhibited by Pipistrellus. This genus has a diploid number varying from 26 to 44, and fundamental numbers from 44 to 60, these findings apparently supporting Koopman's observations. These authors remarked, however, that the karyotype of capensis is more similar to that of Pipistrellus nanus (2N = 36, FN = 50) than to P. kuhlii (2N = 44, FN = 50), Koopman having thought capensis nearer to the kuhlii group than to the pipistrellus group in which he placed nanus. Williams & Mares (loc. cit.) also found, in contrast, that small Eptesicus from the New World (diminutus,furinalis) have the typical 'eptesicoid' karyotype (2N = 50, FN = 48-50), and added that the karyotypic differences between Eptesicus (sensu stricto) and Pipistrellus might prove more useful for separating these genera than other structural features. Our study of the bacula of African 'Eptesicus ' confirms these observations and indicates the isolation of capensis, tenuipinnis and their relatives from Eptesicus sensu stricto (vide infra): Heller & Volleth (1984) also transferred capensis to Pipistrellus, entirely on account of its published karyology. It is interesting to note also that the baculum of P. nanus indicates that this species should be referred to P. (Hypsugo) rather than to P. (Pipistrellus) where Koopman (1975) effectively allocated it. The bacular morphology of capensis, tenuipinnis and their allies suggests strongly that these former groupings of Eptesicus are most closely allied to P. (Hypsugo) as the karyological similarity of capensis to P. nanus indicates. The anterior upper premolar (pm 2 ) is very small, vestigial or absent in P. savii and is very small in most other members of P. (Hypsugo): very rarely it is present in capensis (Wallin, 1969; Hill & Topal, 1973). On the same point, we have been able to examine a specimen (MJS 2846) from Somalia, in the Carnegie Museum of Natural History, which has a small pm 2 on both sides of the jaw, leading to its erstwhile identification as Pipistrellus deserti. The baculum, however, is characteristically that of 'Eptesicus' somalicus, which in fact the specimen represents. Koopman (1975) suggested that Vesperus bicolor Bocage, 1889 ( = Eptesicus bicolor) and Pipistrellus anchietae (Seabra, 1900), both from Angola, may be conspecific, having examined syntypes of both at the British Museum (Natural History). This author thought that bicolor might be a form of 'Eptesicus ' tenuipinnis as Hayman & Hill (1971) suggested, or that it might be based on a specimen of Pipistrellus anchietae with missing anterior upper premolars. Bocage (loc. cit.) says 'pas de trace de la premiere premolaire a la machoire superieure'. Further study of the syntype (BM(NH) 89.5. 1 .3) in London shows it to have a small pm 2 in a recess between c 1 and pm 4 on each side: cranially it agrees exactly with the syntype of anchietae (9 BM(NH) 6. 1 .3. 1) and its baculum is exactly as in that species. Curiously, Bocage states that both original specimens of bicolor are female. The specimen in London is quite clearly listed as a 'Co-type' by Thomas in the relevant accession register. (a) capensis group Tip of baculum flattened, deflected ventrally, sometimes a small sub-apical dorsal projection; braincase flattened, slightly elongate; rostrum not especially broadened; palate long, narrow, interdental palate longer than wide; i 3 1/2 or less the crown area of i 2 . Included taxa: capensis (Fig. 12b, g); probably brunneus, garambae, grandidieri; guineensis (Fig. 12c), matroka (Fig. 12a), melckorum (Fig. 12f); minutus (?) (Fig. 12i); probably rectitragus; somalicus (Fig. 12h); probably vansoni; zuluensis (Fig. 12d). The baculum of brunneus sensu stricto has not been examined. That (Fig. 14b) of a Nigerian specimen (BM(NH) 48.702) collected by I.T. Sanderson and hitherto referred to this species is very similar to that of rendalli (Fig. 12e), with which this example agrees in cranial and ventral 256 J. E. HILL & D. L. HARRISON Dobson he thought to have but one pair of upper incisors. Trouessart (1897) initially followed Dobs~>n (1878) in allocating schlieffenii to Scotozous as a subgenus of Vesperugo, but later (1904) changed this opinion to consider Scotozous a subgenus of Scotophilus. Miller (1907) referred schlieffenii to Scoteinus, although in fact the species does not display the reduction of m| that he considered diagnostic for this genus and which occurs in the Australian species (balstoni, greyii, now incorporated into Scotorepens) that he allocated to it. Miller's view was adopted by Thomas & Wroughton (1908) and in differing ways by many subsequent authors. However, Allen (1911) when describing africanus referred it to the hitherto American genus Nycticeius, commenting on its similarity to N. humeralis and Hollister (1918) remarked that Old World bats usually placed in the genus Scoteinus did not seem to differ generically from the American species of Nycticeius, to which he also referred africanus. Since then africanus has been relegated to subspecific status or synonymy in schlieffenii (Braestrup, 1935; Allen, 1939; Aellen, 1952). Braestrup (loc. cit.) also employed Nycticeius for schlieffenii in preference to Scoteinus, and pointed out that its last upper molar was not reduced in the way that Miller (1907) had described for that genus. This author drew attention to the affinity thus established between the Ethiopian and American faunas, but did not exclude the possibility of convergent evolution from different Pipistrellus-\ike forms. Tate (1942a) maintained Nycticeius and Scoteinus as distinct genera but Simpson (1945) united them, a lead followed by many modern authors who have considered Scoteinus a subgenus of Nycticeius. Thus Ellerman & Morrison-Scott (1951) and Ellerman et al. (1953) referred schlieffeni to Scoteinus as a subgenus of Nycticeius, while Laurie & Hill (1954) listed the Australian species before then allocated to Scoteinus in Scoteanax and Scotorepens as further valid subgenera of Nycticeius. On the other hand, Rosevear (1965) considered Nycticeius and Scoteinus synonymous. Koopman (1965) referred schlieffenii to Scoteinus as a subgenus of Nycticeius but later (in litt. in Hayman & Hill, 1971) revised this opinion to allocate it to Nycticeius (Nycticeius), since then (1978) reinforcing this view. The classification of the Vespertilioninae Earlier classifications of the Vespertilioninae (Miller, 1907; Tate, 19420) rely heavily on the pattern of reduction of the incisor and premolar teeth, chiefly on the presence or absence of the outer upper incisor (i 3 ), of one or both of the first (pm 2 ) or second (pm 3 ) upper premolars, and on the presence or absence of the second (pm 3 ) of the lower premolars, as Tate's 'phyletic' diagrams (loc. cit.) indicate. These dental features have been discussed in more detail above (p. 230): they reflect the degree of shortening that forms an evident trend within the subfamily. When combined with the relative size of one or more of these teeth and the position of the relevant tooth or teeth in the toothrow such factors form an important element in generic identification and diagnosis (cf. Miller, loc. cit.). The many different combinations of incisive and premolar formula in the subfamily (Table 2), the evanescence in some genera of some of the teeth involved, the extreme tendencies towards reduction seen in some such as Pipistrellus, and the variety of positions within the toothrow adopted by i 3 and pm 2 in particular reinforce the conclusion that such features reflect a universal trend that may have occurred more than once within the group and which as a result may not provide a totally reliable yardstick by which relationship may be judged. In addition to these dental features, Tate (1942a) reviewed a number of other characters used in the classification of the subfamily. These include the presence or absence of accessory canine cusps; the form and shape of the braincase and rostrum; the degree of reduction of the zygomata; the structure of the palate, its anterior emargination and accessory anterior and posterior spines; the presence or absence of basial pits; enlargement of the ears and their associated bony structures; the presence or absence of adhesive pads on the thumb or foot; and the nature of other minor structures such as the calcar. These features, however, seem of greater value in the distinction of species and species groups, that is, for infrageneric classification, or for the diagnosis of individual genera. The value of such characters has been discussed at some length by Zima & Horacek (1985) who pointed out that there are grounds for thinking that some of the traditional morphological VESPERTILIONINE SYSTEMATICS 257 characters may not provide unequivocally reliable criteria for the establishment of a classification based on presumed phyletic relationship, and that their taxonomic significance may be limited. They also remarked that such characters may reflect parallelism or convergence, or result from selection pressure rather than relationship. These reasons led them to suggest that the baculum might provide one of several alternative sources of reliable, taxonomically useful criteria based on characters that do not have a direct adaptive significance. The structure of the baculum in the Vespertilioninae suggests some modifications to tribal classification within the subfamily, although clearly other morphological characters need to be given equivalent or greater weight. Provisionally, therefore, we offer an arrangement of the Vespertilioninae in which bacular morphology is used in association with the traditional diagnostic features to suggest possible relationship. This classification is presented in Table 1 . There appear to be two major bacular types in the Vespertilioninae, each with numerous variations as might be expected in such a large and diverse subfamily. A classification that includes a major consideration of bacular morphology shows significant resemblances to earlier arrangements based on traditional and conventional morphological features. However, there are some wide divergences, as for example the seemingly artificial nature of the 'Nycticeini' or the associations of the various genera of big-eared bats. Tate (19420) commented upon the latter and pointed out that very large ears and their associated auditory specialisations in the skull occurred independently in three sections of the subfamily: indeed, if Antrozous and Bauerus are included, these features occur four times in the group. In particular, both Miller (1907) and Tate (loc. cit.) associated Laephotis with Histiotus on cranial and dental morphology but its bacular structure shows a clear affinity with Pipistrellus (Neoromicid) as here recognised. Otonycteris, another big-eared bat, was allied by Tate (loc. cit.) to the 'Nycticeini' but proves to have a baculum much more like those of the plecotine genera. One major bacular type is 'saddle-like' or 'slipper-like' and is exemplified by Myotis and Pizonyx. Their bacula are very similar, emphasising the close relationship that is generally accepted between these genera. The baculum of Lasionycteris is somewhat different in the presence of a lengthened shaft. However, in comparison with the long-shafted bacula found in the Pipistrellini the baculum of Lasionycteris is relatively short, and it retains indications of the more characteristic myotine type in its upraised proximal and distal portions. The occasional presence of a flattened dorsal prominence on its base also recalls the condition found in Idionycteris. The genus, although having some specialised features, is allied firmly to Myotis by Miller (1907) and Tate (\942a). It has slightly hooked upper incisors, i 3 with a slightly caniniform profile as in Myotis', pm 2 is in the line of the toothrow; m 3 is unreduced; pm 2 _ 3 are exactly as in Myotis, much smaller than pm 4 , with pm 3 not removed from the line of the other teeth. Although pm 3 has been lost, this appears to be a specialisation; as Tate (loc. cit.) pointed out, pm 2 _ 3 still agree closely with those of the less specialised species of Myotis not only in relation to each other but also in their proportional size relative to pm 4 . Although associated with Myotis, this genus is considerably specialised in other ways (Miller, loc. cit.) and its bacular structure may well reflect this divergence. Its baculum might be regarded as derived from the more typical myotine structure. Bacula variously reminiscent of the saddle shaped structure found in Myotis occur in a number of other genera. Such bacula characterise Plecotus (including Coryhorhinus), Idionycteris, Barbastella, Rhogeessa, Baeodon, Nycticeius, Otonycteris, Lasiurus, Dasypterus, Antrozous and Bauerus, and possibly may be found in Euderma. Tate (19420) postulated the grouping 'Plecotini' for Plecotus, (Corynorhinus), Idionycteris and Euderma, allying it to the Myotini but not employing the term in a formal taxonomic or systematic sense. Bacular morphology thus lends support to his hypothesis that the plecotine genera should be associated with Myotis. Also, the baculum of Barbastella suggests that it too belongs here: Miller (1907) postulated such a relationship, despite several morphological differences. Rhogeessa, Baeodon, Nycticeius and Otonycteris also seem allied to this grouping. Tate (19420) referred these genera to the 'Nycticeini' with Scotoecus, Scotomanes and Scotophilus on account of their incisive and premolar dentition. However, the bacula of Rhogeessa, Baeodon, Nycticeius and Otonycteris are variants of the saddle-like type; that of Scotoecus is like that of Pipistrellus (Pipistrellus), and the bacula of Scotomanes and Scotophilus are broadly similar to those ofEptesicus and its allies. Lasiurus, Dasypterus, Antrozous and Bauerus 258 J. E. HILL & D. L. HARRISON have further variants of this bacular type, but are quite distinctive on other morphological grounds. The bacula of Antrozous and Bauerus are not at all like that of Otonycteris, with which these genera have been tentatively associated (Pine et al., 1971), nor do their bacula have any significant resemblance to those of Nyctophilus or Pharotis, thus supporting the view (Koopman, 19846, 1985; Breed & Inns, 1985) that these North American genera should not be associated with the Australian Nyctophilus and Pharotis in the subfamily Nyctophilinae. Bacular morphology suggests instead an association with those genera that have the myotine type of baculum, to which the bacula of Antrozous and Bauerus have many resemblances. The bacula of Nyctophilus and Pharotis (Fig. 22a-h) are consistently homogeneous and differ in many ways from those of the genera usually referred to the Vespertilioninae. For the present we would place these two genera in a separate subfamily, the Nyctophilinae, rather than merge them into the Vespertilioninae as is done by Koopman (1984a, 19846, 1985). A further basically triangular and flattened variant of the saddle-like baculum characterises the genera Eptesicus, Vesper tilio (if the pseudobaculum is ignored), la and Histiotus. Miller (1907) remarked that the skull of Vespertilio showed a strong likeness to that of Lasionycteris but that the former was in all respects a typical Eptesicus. Vespertilio and Lasionycteris are separated by marked dental and bacular differences: the bacular morphology of Vespertilio allies it with Eptesicus as Miller suggested. It is perhaps not unreasonable to speculate that Lasionycteris which has a strongly myotine dentition has diverged among the Myotini in the same way as Vespertilio has diverged among the Vespertilionini, the latter genus supporting a long penis either by a centrally situated baculum or perhaps more effectively by the development of a cartilaginous pseudobaculum, this function in Lasionycteris by a short shaft. The genera Tylonycteris and Mimetillus also belong here. The African Glauconycteris has been associated (Ryan, 1966; Koopman, 1971) with the Australian Chalinolobus but their bacula differ widely. Although structurally variable within the genus, the bacula of Glauconycteris are more like the vespertilionine or eptesicine type: those of Chalinolobus are long-shafted and like the bacula of Pipistrellus (Pipistrellus). Finally, the baculum ofScotomanes appears to be a derivative of the saddle-like type, leading to the distinctive baculum of Scotophilus. The genus Pipistrellus seems to stand more or less at the centre of the second major grouping. It has broadly two divisions in bacular terms, one characterised by a long baculum with well developed basal lobes and a relatively long, mostly cylindrical shaft, its tip often bifid or with similar elaboration. The second division includes those species in which the basal lobes are sometimes small or obsolete and which have a shorter, flatter, ventrally fluted shaft, its tip sometimes elaborated into a spatulate or platelet-like structure. These groupings have been used in this study to support subgeneric division of this large genus. The first division includes Pipistrellus (Pipistrellus}, P. (Vespadelus), P. (Perimyotis} and P. (Arielulus}. Reduction and loss of pm 2 occurs in P. (Arielulus) and the tooth is almost invariably absent in P. (Vespadelus}. The second division contains P. (Hypsugo) in which pm 2 may be very small or absent, P. (Neoromicia) from which it is again almost invariably absent, and P. (Falsistrellus}. Although primarily Old World in distribution, both of these divisions are represented in the New World, each by a single species. The Australian P. ( Vespadelus} seems on bacular features to represent P. (Pipistrellus); the wholly African P. (Neoromicia) is apparently similarly related to P. (Hypsugo), of which P. (Falsistrellus} appears to be an eastern representative. The majority of the genera here allocated to the Pipistrellini show strong bacular affinities to Pipistrellus (Pipistrellus): some such as Glischropus and Scotozous have been considered congeneric with Pipistrellus in the past. Besides Glischropus and Scotozous these include Nycticeinops, Scoteanax, Scotorepens, Scotoecus, Nyctalus, Hesperoptenus and Chalinolobus, all with longshafted bacula. Of the remainder, Laephotis in bacular structure is similar to P. (Neoromicia}, while Philetor has a baculum that appears to be an elaboration of the bacular structure found in some of P. (Hypsugo). Tate (1942a) postulated a relationship between Philetor, Tylonycteris and perhaps Mimetillus but the bacula of the first two are totally dissimilar and the structure is apparently absent from Mimetillus: it is very small in Tylonycteris. Hill (1966a) drew attention to the unusual genitalia of Philetor and following Tate's (loc. cit.) suggestion of affinity with Pipistrellus joffrei and VESPERTILIONINE SYSTEMATICS 259 its associates allied Philetor with this group. Unfortunately, excepting for the aberrant species stenopterus the bacula of the stenopterus subgroup (including P.joffrei) of this present study remain unknown. Bacular morphology suggests that the conventional view that Eptesicus and its immediate allies derive from or are closely related to Pipistrellus can be questioned. Cranially and dentally there are many similarities between 'Eptesicus ' as formerly denned and Pipistrellus and as Koopman (1975) has pointed out, the loss of pm 2 enables a species to cross the boundary between the two genera as then understood, a process which in his view might have occurred more than once. Our conclusions do not challenge this opinion: those 'Eptesicus' species in which pm 2 has been found occasionally to occur prove on bacular grounds to be closer to Pipistrellus than to Eptesicus as we understand it, while Pipistrellus as formerly defined has long been known to include some species from which on occasion this 'diagnostic' tooth is absent. Clearly, our findings support Koopman's (loc. cit.) opinion that this process may have occurred several times and indeed may be occurring in some species, but all belong to the one genus, Pipistrellus. As we understand its composition, Eptesicus is now a more restricted genus in which the triangular, flattened baculum is basically closer in structure to the saddle-like grouping than to the long-shafted group, although some Eptesicus do indeed have bacula that suggest the beginnings of basal lobulation or of a very short shaft. We suggest therefore that in bacular terms the Vespertilionini to which we refer Eptesicus may represent a transitional stage between the saddlelike baculum and the predominantly basally lobed and long-shafted type. Tylonycteris and Glauconycteris also show this tendency. Dental reduction proceeds throughout both of the major bacular groups. In the grouping with broadly myotine or saddle-like bacula the dentition varies in number of teeth from a total of 38 (Myotis, Pizonyx) through 36 (Lasionycteris, Plecotus and allies), 34 (Barbastella, Eptesicus and allies), 32 (Lasiurus), 30 (Dasypterus, Rhogeessa, Baeodon, Nycticeius, Otonycteris, Scotomanes, Scotophilus) to 28 (Antrozous, Bauerus}. In the second of the two major bacular groups, dental reduction varies from Eudiscopus with a total of 36 teeth (its association here is presumed) through 34 (Pipistrellus, Glischropus, Scotozous, Nyctalus, Chalinolobus), 32 (Laephotis, Philetor, Hesperoptenus) to 30 (Nycticeinops, Scoteanax, Scotorepens, Scotoecus). Thus this trend occurs concurrently in the two major groupings, taking the same form in each by increasing the size and bulk of i 2 , the reduction, transposition and loss of i 3 , and the progressive reduction, transposition and loss of pm 2 , pm 3 and pm 2 . Zoogeographical considerations The saddle-shaped or slipper-like baculum characteristic of the Myotini, Plecotini, Lasiurini and Antrozoini as here understood is cosmopolitan in but one genus, Myotis. It occurs in one Holarctic genus, Plecotus, in one Palaearctic genus, Barbastella, itself probably closely related to Plecotus, and in one other Old World genus, Otonycteris, that occurs in southwestern Asia and northern Africa. Otherwise this bacular type is limited to the New World. Lasionycteris, exclusively North American, has a baculum apparently derived from this type, as does Nycticeius, also North American, although in this genus the baculum is considerably modified to the extent that Hamilton (1949) commented upon its unique character among the genera that he had examined. Thus although the saddle-shaped baculum or its derivatives is represented about equally in number of species in the Old and New Worlds, genera with bacula of this type predominate in the latter, its extension into the Old World being primarily through the many species of Myotis, with a lesser contribution from Plecotus, Barbastella, and Otonycteris. A further variety of this bacular type is found in the Vespertilionini, that is, in Eptesicus and its close relatives. In these, the baculum is less strongly saddle-shaped or slipper-like, flatter, and often more triangular in outline. This bacular type is primarily Old World in numbers of genera and species, only Eptesicus among Old World genera extending to the New World where there is a closely related genus, Histiotus. In the Old World, Vespertilio is also closely related to Eptesicus. Another Old World genus, la, is a giant representative of this same bacular type. The southeastern 260 J. E. HILL & D. L. HARRISON Asian Tylonycteris and the African Glauconycteris have bacula that are modified variants of this type: Mimetillus, in which no baculum has been found, also appears to belong here. Two further Old World genera, Scotomanes and Scotophilus, also have bacula that are similar in many respects to the vespertilionine type. The shafted or long-shafted bacular type is confined almost exclusively to the Old World, and is represented in the New World by no more than two species of Pipistrellus in the Nearctic region, one of these with a highly modified baculum. This bacular type is restricted to the Pipistrellini and within that grouping to those genera that for the most part can be shown on other grounds to cluster around Pipistrellus. Indeed, some such as Scotozous, Glischropus, Scoteanax, Scotorepens and perhaps even Nyctalus might on bacular grounds be regarded as subgenera of this widespread genus. In a reduced form this bacular type appears in two of the subgenera of Pipistrellus, P. (Perimyotis) and P. (Arielulus). Widespread in the Palaearctic region and in southeastern Asia, this bacular type is represented in Australia by five distinct groupings: Pipistrellus (Pipistrellus}, P. ( Vespadelus), Scoteanax, Scotorepens, and Chalinolobus. This type of baculum also occurs in Africa among Pipistrellus kuhlii and its associates, which might in fact be considered to warrant recognition as a further subgenus of Pipistrellus. A further variant of the shafted bacular type is found in Pipistrellus (Hypsugo) and P. (Falsistrellus). In these the shaft is shorter and is ventrally fluted, often with expansion of the tip. Pipistrellus (Hypsugo) is confined chiefly to Asia and Africa, where in the latter region it appears to be closely associated with P. (Neoromicia) in which pm 2 is generally lost. Thus as in Australia where P. (Vespadelus) in which pm 2 is also generally absent appears to derive from P. (Pipistrellus), so in Africa P. (Neoromicia) is apparently similarly related to P. (Hypsugo). Of the two North American pipistrelles, P. subflavus has a reduced form of the P. (Pipistrellus) baculum, the shaft very short and stubby: this species has a myotine tragus and has been considered (Menu, 1984) to have a myotine dentition. However, on the balance of features it appears to be clearly referable to Pipistrellus and indeed to be cranially and dentally close to P. (Pipistrellus), which apparently it represents in North America. There do not appear to be sufficient grounds to justify its generic separation from Pipistrellus as has been recently effected (Menu, loc. cit.), although subgeneric recognition within that genus seems appropriate. The second North American species of Pipistrellus, P. Hesperus, should evidently be referred to P. (Hypsugo) with which it has close bacular and dental similarities, although recently generic separation (Horacek & Hanak, 19850, b, 1985-1986) has been proposed for it. Finally, P. (Falsistrellus) is restricted to southeastern Asia, Australasia and Tasmania: the deeply ventrally fluted baculum of this subgenus, lacking basal and distal modification but massive and substantial appears to be an extreme of the P. (Hypsugo) type: possibly P. (Falsistrellus) represents P. (Hypsugo) which seems to be linked to it by several of its Asian species. One corollary of the removal of the African capensis and tenuipinnis groups of 'Eptesicus' to Pipistrellus, and of the similar transfer of the Australian species formerly referred to 'Eptesicus' is that in the Old World Eptesicus now becomes primarily Palaearctic, with outliers, perhaps all closely connected to E. serotinus, in Africa while in the New World it extends over both North and South America. In southeastern Asia the genus becomes restricted to no further east than southern Thailand, the former enormous hiatus in its distribution between this part of southern Asia and Australia having been removed. Conclusions (1) The current classification of the Vespertilioninae is based chiefly on adaptive characters with considerable emphasis on facial shortening and concomitant dental reduction and loss. Several authors have drawn attention to the deficiencies and dangers of any classification that relies heavily on such features. A review of bacular morphology within the subfamily suggests that this structure provides indications of relationship that in many respects support the existing classification but which also indicate several changes to the current arrangement. In particular, bacular morphology suggests a number of major and minor changes in the systematics of the nominal genera Pipistrellus VESPERTILIONINE SYSTEMATICS 261 and Eptesicus, separated hitherto only by dental formula, itself subject to variation in both 'genera' as they are currently understood. (2) The presence or absence of the anterior upper premolar (pm 2 ) in Pipistrellus and Eptesicus, used formerly as their principal diagnostic character, has little taxonomic significance. The tooth is variable in Pipistrellus as here understood, being reduced or lost in three of its subgenera, and is absent from Eptesicus as we envisage it. (3) Bacular morphology in Pipistrellus and Eptesicus provides groupings that largely agree in species content with those proposed by earlier authors such as Tate (19420) and Koopman (1973, 1975) although in basing their studies on 'conventional' morphological characters neither considered these genera in their entirety. The bacular morphology of 'Eptesicus' as it is currently understood provides a clear indication that as such it is not a natural group, but that three species aggregations, the Australian pumilus group and the African capensis and tenuipinnis groups, should be transferred to Pipistrellus. (4) It has been possible to recognise and define subgenera for the major species groups in both Pipistrellus and Eptesicus and to suggest possible relationships between them. One subgenus is described as new as Pipistrellus (Arielulus) for P. circumdatus and its allies. (5) There appear to be clear links between certain of the pipistrelline subgenera: Pipistrellus ( Vespadelus} in Australia seems to represent P. (Pipistrellus) in bacular terms while P. (Hypsugo) is apparently represented in Indo-Australia by P. (Falsistrellus) and is related to the African P. (Neoromicid). Although the features of the two Nearctic species of Pipistrellus have been thought to justify their recognition in separate, individual genera we consider that the characters of one (subflavus) merit no more than subgeneric status as the sole species of P. (Perimyotis), which itself perhaps represents P. (Pipistrellus), while the other (Hesperus) is perhaps more appropriately referred to P. (Hypsugo). (6) The examination of bacula in Pipistrellus has suggested that some taxa hitherto ranked as subspecies, for example abramus, paterculus or helios, might in fact be distinct species. (7) As we now understand the species content of Pipistrellus and Eptesicus the former remains primarily an Old World genus where it is widespread and diverse in the tropics and subtropics, extending into the temperate zones and just to North America. In contrast, our concept of Eptesicus limits this genus to the New World and in the Old World primarily to the Palaearctic, with outlying representatives in Africa. It does not extend significantly into Australasia. (8) Bacular morphology suggests the informal recognition of two major groupings within the subfamily Vespertilioninae. The first includes the Myotini, Plecotini and Lasiurini; Antrozous and Baeurus, which in bacular terms have no relation to Nyctophilus and Pharotis (the Nyctophilinae); the Scotophilini to include Scotomanes and Scotophilus; and finally the Vespertilionini, here reduced in content to include Eptesicus and its close relatives Histiotus, la and Vespertilio, with Tylonycteris, Mimetillus and Glauconycteris. (9) The second grouping consists of Pipistrellus and those genera which cluster round it. All with the possible exception of Philetor appear to relate quite closely in bacular terms to one or other of the subgenera that we recognise in Pipistrellus, principally to P. (Pipistrellus). Laephotis, formerly considered related to Histiotus, is instead in bacular terms closely associated with P. (Neoromicid). The bacula of Chalinolobus and Glauconycteris are widely dissimilar athough these genera have been closely allied in the past; Chalinolobus is of the pipistrelline type while the baculum of Glauconycteris apparently associates it more appropriately with Eptesicus and its allies. (10) Bacular morphology provides clear indications that the 'Nycticeini' of Tate (\942a) and Koopman (1984, 1985) is not a natural group, its constituent members despite cranial and dental similarities having widely different bacula. Thus Rhogeessa, Baeodon, Nycticeius sensu stricto, and Otonycteris have been here allied to the plecotine bats on bacular grounds, while Scoteanax, Scotorepens and Scotoecus are quite clearly associates in bacular terms of Pipistrellus. 'Nycticeius', at one time thought to include the Australian Scoteanax and Scotorepens as well as its North 262 J. E. HILL & D. L. HARRISON American type species humeralis and the African schlieffenii, has recently been restricted only to the American and African forms. These prove to have widely different bacula; humeralis has been associated with the plecotine bats on this account, while generic status has been accorded to schlieffenii with the proposal of a new generic name, Nycticeinops. (11) The two broad bacular types that we discern in the subfamily Vespertilioninae have definite geographical patterns: the saddle-like baculum and its variants that characterise the first group noted above is primarily New World and Palaearctic, extending less significantly into the Old World tropics or Australasia, while the shafted baculum of the second group is chiefly confined to the Old World. Addendum A phenetic analysis of the relationships of selected vespertilionine species (chiefly those currently referred to Pipistrellus and Eptesicus) by Horacek & Hanak (1985-1986) appeared while this paper was in press. These authors provided definitions of Pipistrellus, Hypsugo (which they considered generically valid) and Eptesicus, based on the morphology of the penis and baculum, the upper molars, the basisphenoid pits, the pelvic girdle, and the tibia, tail and epiblema. Horacek & Hanak suggested that the classification of pipistrelloid bats might be clarified by the recognition of additional subgenera or genera for those species or species groups that do not conform precisely with those that they included within these three generic groupings. To some extent such recognition is provided in several instances by the classification here proposed and although some major differences exist between the informal assessments and species groups of Horacek & Hanak and the formal arrangement put forward in this paper there is nevertheless a broad measure of agreement. Horacek & Hanak did not attempt any classification of the Vespertilioninae as a whole, but 'Nycticeius' schlieffenii, here considered to represent a distinct monospecific genus (Nycticeinops gen. nov.) was thought by these authors to be referable either to Eptesicus ( Rhyneptesicus), or possibly to justify the establishment of a new subgenus within Eptesicus. References Acloque, A. 1899. Faune de France, contenant la description des especes indigenes disposees en tableaux analytiques et illustre de figures representant les types caracteristiques des genres. I. Mammiferes, Oiseaux, Poissons, Reptiles, Batraciens, Protochordes. Paris: Librairie J.-B. Balliere et Fils. Aellen, V. 1952. Contribution a 1'etude des Chiropteres du Cameroun. Memoires de la Societe Neuchateloise des Sciences Naturelles, Neuchatel 8: 1-121. - 1957. Les chiropteres africains du Musee Zoologique de Strasbourg. Revue Suisse de Zoologie, Geneve 64:189-214. 1959. Chiropteres nouveaux d'Afrique. Archives des Sciences, Geneve, 12: 217-235. Agrawal, V. C. & Sinha, Y. P. 1973. Studies on the bacula of some Oriental bats. Anatomischer Anzeiger, Jena 133: 180192, 4 figs. Allen, G. M. 1908. Notes on Chiroptera. Bulletin of the Museum of Comparative Zoology, Harvard 52: 25-62. - 1911. Bats from British East Africa. Bulletin of the Museum of Comparative Zoology, Harvard 54: 321-331. - 1914. Mammals from the Blue Nile Valley. Bulletin of the Museum of Comparative Zoology, Harvard 58: 305-357. 3 figs 1939. A checklist of African mammals. Bulletin of the Museum of Comparative Zoology, Harvard 83: 1-763. Allen, H. 1891. Change of name of a genus of bats. Proceedings of the Philadelphia Academy of Natural Sciences 43: 466. Anderson, K. 1912. Catalogue of the Chiroptera in the collection of the British Museum. 2nd ed. Vol. I: Megachiroptera. London: Trustees of British Museum. Atallah, S. I. 1970. Bats of the genus Myotis (Family Vespertilionidae) from Lebanon. Occasional Papers University of Connecticut , Storrs, (Biological Sciences Series) 1: 205-212, 2 figs, 1 tab. VESPERTILIONINE SYSTEMATICS 263 Baagee, H. J. 1973. Taxonomy of two sibling species of bats in Scandinavia, Myotis mystacinus and Myotis brandti (Chiroptera). Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening i Kjobenhavn 136: 191-216, 17 figs, 2 tabs. Bhatnagar, K. P. 1967. Bacula of some Indian Megachiroptera. Journal of Mammalogy, Baltimore 48: 494-497, 3 fi gs . Bianchi, V. 1917. Notes preliminaries sur les chauve-souris ou Chiropteres de la Russie. Annuaire du Musee Zoologique de I'Academie Imperiale des Sciences, St Petersburg (Leningrad) (1916) 21: LXXIII-LXXXII. (In Russian). Blainville, H. M. D. de 1840. Osteographie ou description iconographique comparee du squelette et du systeme dentaire des Mammiferes, recents et fossiles pour servir de base a la zoologie et a la geologie. Paris: J. B. Balliere et Fils. Blyth, E. 1840. The Mammalia, Birds and Reptiles. In Cuvier, G. [L.C.F.D.], The Animal Kingdom, arranged after its organisation, forming a natural history of animals, and an introduction to comparative anatomy. London: W. S. Orr & Co. Bocage, J. V. B. du 1889. Chiropteres africains nouveaux, rares ou peu connus. Jornal de Sciencias Mathematicas, Physicas e Naturaes, Lisboa (2), 1: 1-7, 2 figs. Bonaparte, C. L. J. L. 1832-1841. Iconographia della Fauna Italica per le quattro classi degli Animali Vertebrati. Roma: Tipografia Salviucci. Braestrup, F. W. 1935. Report on mammals collected by Mr Harry Madsen during Professor O. Olufsen's Expedition to French Sudan and Nigeria in the years 1927-28. Vidensk Meddelelser fra Dansk Naturhistorisk Forening i Kjobenhavn 99: 73-130. Breed, W. G. & Inns, R. W. 1985. Variation in sperm morphology of Australian Vespertilionidae and its possible phylogenetic significance. Mammalia, Paris 49: 105-108, 2 pis. Brown, R. E. 1967. Bacula of some New World molossid bats. Mammalia, Paris 31: 645-667, 51 figs. , Genoways, H. H. & Jones, J. K. Jr. 1971. Bacula of some Neotropical bats. Mammalia, Paris 35: 456-464, 1 fig. Burt, W. H. 1960. Bacula of North American mammals. Miscellaneous Publications, Museum of Zoology, University of Michigan, Ann Arbor No. 113: 1-75, 25 pis. Chaine, J. 1926. L'os penien, etude descriptive et comparative. Actes de la Societe Linneenne de Bordeaux 78: 5-195, 133 figs. Chasen, F. N. 1940. A Handlist of Malaysian mammals. Bulletin of the Raffles Museum, Singapore No. 15; i-xx, 1-209, map. Churchill, S. K., Hall, L. S. & Helman, P. M. 1984. Observations on long-eared bats (Vespertilionidae: Nyctophilus) from northern Australia. Australian Mammalogy 7: 17-28, 6 figs, 3 tabs. Corbet, G. B. 1964. The grey long-eared bat Plecotus austriacus in England and the Channel Islands. Proceedings of the Zoological Society of London 143: 51 1-515, 2 tabs. & Hill, J. E. 1980. A world list of mammalian species. 1st ed. London/Ithaca: British Museum (Natural History)/Comstock Publishing Associates, Cornell University Press. 1986. A world list of mammalian species. 2nd ed. New York/London: Facts on File/British Museum (Natural History). Daubenton, E. L. 1 760. Histoire naturelle generate et particuliere avec la description du Cabinet du Roi. Tome 8. Paris: Imprimerie Royale. Davis, D. D. 1947. The bacula of some fruit bats (Pteropus). Fieldiana: Zoology, Chicago 31: 125-131, 2 figs. Davis, W. H. & Rippy, C. L. 1968. Distribution of Myotis lucifugus and Myotis austroriparius in the southeastern United States. Journal of Mammalogy, Baltimore 49: 1 13-1 17, 2 figs. De Blase, A. F. 1980. The bats of Iran: systematics, distribution, ecology. Fieldiana: Zoology, Chicago N.S. No. 4 (Pub. 1307): i-xvii, 1^24, 180 figs. Dobson, G. E. 1 87 1 . Notes on nine new species of Indian and Indo-Chinese Vespertilionidae, with remarks on the synonymy and classification of some other species of the same family. Proceedings of the Asiatic Society of Bengal, Calcutta 210-215. - 1876. Monograph of the Asiatic Chiroptera and Catalogue of the species of bats in the collection of the Indian Museum, Calcutta. London: Trustees of Indian Museum. 1878. Catalogue of the Chiroptera in the collection of the British Museum. London: Trustees of British Museum. I Herman, J. R. & Morrison-Scott, T. C. S. 1951. Checklist of Palaearctic and Indian mammals 1758-1946. London: Trustees of British Museum (Natural History). , & I layman, R. W. 1953. Southern African mammals 1758-1951: a ^classification. London: Trustees of British Museum (Natural History). 264 J. E. HILL & D. L. HARRISON Ercolani, G. B. 1868. Dei Tessuti e degli organ! erettili. Memorie dell 'Accademia delle Scienze dell'Instituto di Bologna (2), 8: 281-362, 10 pis. Fairon, J. 1980. Deux nouvelles especes de Cheiropteres pour la fauna du Massif de 1'Air (Niger); Otonycteris hemprichi Peters, 1 859 et Pipistrellus nanus (Peters 1 852). Bulletin de I'lnstitut Royal des Sciences Naturelles de Belgique, Bruxelles 52 (17): 1-7, 2 figs, 2 photos, 1 tab. Fischer, J. B. 1829. Synopsis mammalium. (With addenda et emendanda, index, corrigenda). Stuttgardiae. Fitzinger, L. J. 1870. Kritische Durchsicht der Ordnung der Flatterthiere oder Handfliigler (Chiroptera). Familie der Fledermause (Vespertiliones). V. Abtheilung. Sitzungberichte der [ Kaiser lichen J Akademie der Wissenschaften. Mathematisch-Naturwissenschaftliche Classe, Wien 62: 353-438. Gaisler, J., Madkour, G. & Pelikan,. J. 1972. On the bats of Egypt. Pfirodovedne prdce iistavu Ceskoslovenske Akademie Ved v Brne (Acta Scientiarum Naturalium Akademiae Scientarium Bohemoslovaceae Brno, N.S. 6 (8): 1-40, 5 figs, 4 pis, 20 tabs. Genoways, H. H. & Jones, J. K., Jr. 1969. Taxonomic status of certain long-eared bats (genus Myotis) from the southwestern United States and Mexico. The Southwestern Naturalist, Dallas 14: 1-13, 5 figs, 1 tab. Gerhardt, U. 1905. Morphologische und biologische Studien iiber die Kopulationsorgane der Saugetiere. Jenaische Zeitschrift fur Naturwissenschaft, Jena 39: 43-1 18, 1 pi. Gilbert, T. 1892. Das Os priapi der Saugethiere. Morphologisches Jahrbuch, Leipzig 18: 805-831, 1 pi. Gray, J. E. 1838. A revision of the genera of bats (Vespertilionidae) and the description of some new genera and species. Magazine of Zoology and Botany, Edinburgh & Dublin 2: 483-505. - 1866. Synopsis of the genera of Vespertilionidae and Noctilionidae. Annals and Magazine of Natural History, including Zoology, Botany and Geology, London (3), 17: 89-93. Hamilton, W. J. 1949. The bacula of some North American vespertilionid bats. Journal of Mammalogy, Baltimore 30: 97-1 02, 1 pi. Hanak, V. 1965. Zur systematik der Bartfledermaus Myotis mystacinus Kuhl, 18 19 und iiber das Vorkommen von Myotis ikonnikovi Ognev, 1912 in Europa. Vestnik Ceskoslovenske Spolecnosti Zoologicke, Praha 29: 353-367, 9 figs, 3 tabs. 1970. Notes on the distribution and systematics of Myotis mystacinus Kuhl, 1819. In Proceedings of the Second International Bat Research Conference, Amsterdam, 1970. Bijdragen tot de Dierkunde , An sterdam & Leiden 40 (1): 40-44, 8 figs, 1 tab. 1971. Myotis brandtii (Eversmann, 1845) (Vespertilionidae, Chiroptera) in der Tschechoslovakei. Vestnik Ceskoslovenske Spolecnosti Zoologicke, Praha 35: 175-185, 7 figs, 1 tab. Harrison, D. L. 1960. A new species of pipistrelle bat (Chiroptera: Pipistrellus) from south Israel. Durban Museum Novitates 5(19: 261-267, 2 figs, 2 pis. - 1982. Observations on some rare Arabian Pipistrellus (Chiroptera: Vespertilionidae) with special reference to the external male genitalia. Banner Zoologische Beitrdge 33: 187-190, 2 figs. & Brownlow, I. P. 1978. A comparative study of the baculum in bats of the genus Scotophilus (Chiroptera: Vespertilionidae) Mammalia, Paris 42: 123-130, 7 figs. H ay man. R. W. & Hill, J. E. 1 97 1 . 2. Chiroptera. In Meester, W. & Setzer, H. W. [Eds] The mammals of Africa. An identification manual. Washington: Smithsonian Press. Heller, K-G. & Volleth, M. 1984. Taxonomic position of 'Pipistrellus societatis' Hill, 1972 and the karyological characteristics of the genus Eptesicus (Chiroptera: Vespertilionidae). Zeitschrift fur Zoologische Systematik und Evolutionsforschung, Frankfurt am Main 22 (1): 65-77, 5 figs, 1 tab. Heuglin, M. T. von 1877. Reise in Nordost-Africa. Schilderungen aus dem Gebiete der Beni Amer undHabab, nebst zoologischen Skizzen undeinem Fuhrerfur Jagdreisende. Braunschweig: George Westermann. Hill, J. E. 1966a. A review of the genus Philetor (Chiroptera: Vespertilionidae). Bulletin of the British Museum (Natural History) , (Zoology), London 14: 375-387, 3 figs. - 19666. The status of Pipistrellus regulus Thomas (Chiroptera: Vespertilionidae) Mammalia, Paris 30: 302-307. - 1969. The generic status of Glischropus rosseti Oey, 1951 (Chiroptera: Vespertilionidae). Mammalia, Paris 33: 133-1 39. - 1971. The status of Vespertilio brachypterus Temminck, 1840 (Chiroptera: Vespertilionidae). Zoologische Mededelingen, Leiden 45: 139-146. 1972. The Gunong Benom Expedition 1967. 4. New records of Malayan bats, with taxonomic notes and the description of a new Pipistrellus. Bulletin of the British Museum (Natural History), (Zoology), London 23: 2 1-42, 3 tabs. 1974. A review of Scotoecus Thomas, 1901 (Chiroptera: Vespertilionidae). Bulletin of the British Museum (Natural History), (Zoology), London 27: 167-188, 4 figs, 1 tab. 1976. Bats referred to Hesperoptenus Peters, 1869 (Chiroptera: Vespertilionidae) with the description of a new subgenus. Bulletin of the British Museum (Natural History) , (Zoology), London 30: 1-28, 5 figs, 4 pis. VESPERTILIONINE SYSTEMATICS 265 - 1983. Bats (Mammalia: Chiroptera) from Indo-Australia. Bulletin of the British Museum (Natural History), (Zoology), London 45: 103-208, 14 tabs. & Francis, C. M. 1984. New bats (Mammalia: Chiroptera) and new records of bats from Borneo and Malaya. Bulletin of the British Museum (Natural History), (Zoology), London 47: 305-329, 2 figs, 4 tabs. & Topal, G. 1973. The affinities of Pipistrellus ridleyi Thomas, 1898 and Glischropus rosseti Oey, 1951 (Chiroptera: Vespertilionidae). Bulletin of the British Museum (Natural History), (Zoology), London 24: 447^56. Hollister, N. 1918. East African mammals in the United States National Museum. Part I. Insectivora, Chiroptera and Carnivora. Bulletin of the United States National Museum, Washington No. 99: 1-194, 3 figs, 55 pis. Honacki, J. H., Kinntan, K. E. & Koeppl, J. W. [Eds] 1982. Mammal species of the world: a taxonomic and geographic reference. Lawrence, Kansas: Allen Press/Association of Systematic Collections. Horacek, I. & Hanak, V. 1985a. Generic status of Pipistrellus savii (Bonaparte, 1837) and remarks on systematics of the genus Pipistrellus. Abstracts Seventh International Bat Research Conference Third European Bat Research Symposium. Joint Meeting. Aberdeen: University of Aberdeen. [Unpaginated]. [August]. , - 19856. Generic status of Pipistrellus savii (Bonaparte, 1837) and remarks on systematics of the genus Pipistrellus. Bat Research News, Potsdam, New York 26: 62. [November]. 1985-1986. Generic status of Pipistrellus savii and comments on classification of the genus Pipistrellus (Chiroptera, Vespertilionidae). Myotis, Bonn 23-24: 1 1-16, 4 figs. Ibanez, C. & Fernandez, R. 1986. Systematic status of the long-eared bat Plecotus teneriffae Barret-Hamilton, 1907 (Chiroptera; Vespertilionidae). Sdugetierkundliche Mitteilungen, Miinchen (1985), 32: 143-149, 2 figs, 2 tabs. & Valverde, J. A. 1985. Taxonomic status of Eptesicus platyops (Thomas, 1901) (Chiroptera, Vespertilionidae). Zeitschrift fur Sdugetierkunde, Hamburg & Berlin 50: 241-242. Imaizumi, Y. 1959. A new bat of the " Pipistrellus javanicus" group from Japan. Bulletin of the National Science Museum, Tokyo No. 45, (N.S. 4): 363-371, 6 figs, 1 pi, 3 tabs. Iredale, T. & Troughton, E. le G. 1934. A check-list of the mammals recorded from Australia. Memoir of the Australian Museum, Sydney No. 6: i-xi, 1-122. Kaup, J. J. 1829. Skizzirte Entwicklungs-Geschichte und natiirliches System der Europdischen Theirwelt. Erster Theil, welcher die Vogelsaugethiere und Vogel, nebst Andeutung der Entstehung der letzteren aus Amphibien enthdlt. Darmstadt & Leipzig: Commision die Carl Wilhelm Leste. Khajuria, H. 1979. Studies on the bats (Chiroptera: Mammalia) of M.P., India. Pt. I. (Families Pteropidae, Rhinopomatidae and Embalonuridae [sic]). (Taxonomical and ecological studies on bats of Jabalpur District Madhya Pradesh, India. Part I. Families Pteropidae, Rhinopomatidae and Embalonuridae [sic]). Records of the Zoological Survey of India Miscellaneous Publications Occasional Papers, Delhi No. 13: 1-59, 6 figs, 7 pis, 8 tabs. - 1980. Taxonomical and ecological studies on bats of Jabalpur District Madhya Pradesh, India. Pt. II. (Families Megadermatidae, Rhinolophidae and Vespertilionidae). Records of the Zoological Survey of India Miscellaneous Publications Occasional Papers, Delhi No. 19: 1-73, 8 pis, 8 tabs. 1982. External genitalia and bacula of some central Indian Microchiroptera. Sdugetierkundliche Mitteilungen, Stuttgart 30: 287-295, 5 figs. Kitchener, D. J. 1976. Eptesicus douglasi, a new vespertilionid bat from Kimberley, Western Australia. Records of the Western Australian Museum, Perth 4: 295-301, 1 fig., 2 tabs. & Caputi, N. 1985. Systematic revision of Australian Scoteanax and Scotorepens (Chiroptera: Vespertilionidae) with remarks on relationships to other Nycticeiini. Records of the Western Australian Museum, Perth 12: 85-146, 19 figs, 3 tabs. & Jones, B. 1986. Revision of Australo-Papuan Pipistrellus and Falsistrellus (Microchiroptera: Vespertilionidae). Records of the Western Australian Museum, Perth, 12: 435-495, 20 figs, 3 tabs, appendix. Kolenati, F. A. 1856. Europa's Chiroptern. 1. Synopsis der Europaischen Chiroptera. Allgemeine deutsche Naturhistorische Zeitung (New Folge), Dresden & Leipzig 2: 121-133. [F. A.] 1858. Eine neue osterreichische Fledermaus. Sitzungberichte der [Kaiser lichen] Akademie der Wissenschaften. Mathematisch-Naturwissenschaftlichen Classe, Wien 29: 250^256. Koopman, K. F. 1965. Status of forms described or recorded by J. A. Allen in The American Museum Congo Expedition Collection of Bats'. American Museum Novitates, New York No. 2219: 1-34. - 1971. Taxonomic notes on Chalinolobus and Glauconycteris (Chiroptera, Vespertilionidae). American Museum Novitates, New York No. 2451: 1-10, 1 fig. 1973. Systematics of Indo-Australian pipistrelles. Periodicum Biologorum, Zagreb 75: 1 13-1 16, 3 figs. 272 J. E. HILL & D. L. HARRISON BM(NH) 27.12.1.37 Tarn Dao, Tonkin, Vietnam, 3000 ft (No. 41 1 of Thomas, 1928a). BM(NH) 28.7. 1 .20 Phu Qui, Annam, Vietnam, 100 ft (No. 866 of Thomas, 19286, who identified the specimen as P. coromandra tramatus, but with a longer baculum than those previously examined). BM(NH) 83.76 Silau Silau Trail, Mount Kinabalu, Sabah, Borneo. (Fig. lOe) Pipistrellus paterculus BM(NH) 14.7.8.62 Pyaunggaung, N Shan States, Burma, 2794 ft. BM(NH) 14.7.19.241 Kyauk Myaung, Irrawaddy, W Burma. BM(NH) 14.7.19.242 Mount Popa, Upper Burma (Holotype). (Fig. 3c) BM(NH) 14.7.19.240 Mandalay, Burma. Pipistrellus angulatus BM(NH) 67.2125 Schoolmaster's House, Nuhu, Guadalcanal I, Solomon Is (ponceleti). (Fig. 4d) Pipistrellus collinus BM(NH) 50.983 Baiyanka, Purari-Ramu Divide, SE Bismarck Range, Papua New Guinea. (Fig. 4b) Pipistrellus coromandra BM(NH) 32. 1 1 . 1 .7 Nam Tamai, Upper Burma. BM(NH) 50.478 Ningma, Upper Burma. BM(NH) 76.1263 Sumka Uma, Upper Burma. HZM 1.7317, HZM 2.73 18 Near Mirzapur, India. (Fig. 7c, HZM 2.7318) HZM 4.7320 Dalatpur, near Mirzapur, India. BM(NH) 4.6.8.1 Annam, Vietnam (tramatus). (Fig. 7b) BM(NH) 27.12.1.40 Bac-kan, Tonkin, Vietnam (tramatus) (Original No. 444, seen by Thomas, 1928a). Pipistrellus mimus BM(NH) 98.5.5.20 Dangs, Bombay, India. HZM 1 .10456 Vikas Vidyalaya, near Ranchi, Bihar, India. (Fig. 7g) Pipistrellus murrayi BM(NH) 99.8.6.34 Christmas I, Indian Ocean (Holotype). (Fig. 4c) BM(NH) 9.1.16.7 Flying Fish Cove, Christmas I, Indian Ocean. Pipistrellus papuanus BM(NH) 22.2.2.3 Fredrik Hendrik I, Irian Jaya. (Fig. 2c) BM(NH) 34.1.14.8 Kokoda, Papua New Guinea. Pipistrellus tenuis BM(NH) 85.912 Coast of Sabah, Borneo (nitidus). (Fig. 9d) Pipistrellus ceylonicus BM(NH) 95.6.12.1 Pundibiya, India. BM(NH) 2.4.2.8 Astoli, Belgoum, India. (Fig. 7d) BM(NH) 1 1.4.5.5 Lanje, Konkan, India. BM(NH) 13.9.8.102 Gujerat, India. BM(NH) 9.1.4.73 Mangalore, Malabar Coast, India (Holotype indicus) BM(NH) 4.6.8.7-8 Tonkin, Vietnam (raptor). (Fig. 3d, BM(NH) 4.6.8.7 Holotype). Pipistrellus crassulus BM(NH) 4.2.8.1 Efulen, Cameroun (Holotype). (Fig. 7e) Pipistrellus nanulus BM(NH) 4.2.8.8 Efulen, Cameroun (Holotype). (Fig. If) BM(NH) 79.508 South Nimba, Liberia. Pipistrellus rueppellii BM(NH) 68.12.22.3 Zanzibar (Holotype pulcher). (Fig. lOa) BM(NH) 99.9.9.20 Egypt. BM(NH) - - Uganda. (Fig. lOb) HZM 3.3170 Kabompo Boma, Zambia. HZM 7. 12109 Suez, Egypt. Pipistrellus deserti BM(NH) 79.987 Hoggar Plateau, Algeria. (Fig. 5c) NMW 27503 (?) Upper Egypt. VESPERTILIONINE SYSTEMATICS 273 Pipistrellus kuhlii BM(NH) 92.9.9.25 Upper Egypt. BM(NH) Argostoli, Cephaloni, Greece. (Fig. 5a) BM(NH) 63.335 Sangha, Malya Khola, E Nepal. HZM 5. 1 1607 Horefto, near Volos, Greece. HZM 11.1016 Rapallo, N Italy. HZM 138.4563 Yal bu Hillal, Batinah, Oman. HZM 154.4619 Saham, Batinah, Oman. HZM 203.7232 Dig Dagga, Ras al Khaima, United Arab Republic. HZM 218.7402 Benghazi, Libya. HZM 227.91 10 Kapsowat, Marakwat, Kenya. Pipistrellus maderensis BM(NH) 86.528 Madeira. (Fig. 5b) Pipistrellus rusticus BM(NH) 35.9.1.108 Okavango-Omatako Junction, Grootfontein District, Namibia. BM(NH) 79.1731 Oli River, Borgu G.R., Nigeria. (Fig. 6c) HZM 4.3285 Sentinel Ranch, River Limpopo, Zimbabwe. (Fig. 5d) Pipistrellus ( Vespadelus} Pipistrellus pumilus BM(NH) 70.1093 E Bonithon Range, C Australia 2342'S, 12902'E, 1400 ft. BM(NH) 71.1497 Westwood, near Rockhampton, Queensland, Australia. (Fig. 12k) Pipistrellus (Perimyotis) Pipistrellus subflavus HZM 1.2422 Big Wyandotte Cave, Crawford County, Indiana, USA. (Fig. 2d) Pipistrellus (Hypsugo) Pipistrellus anchietae BM(NH) 69.1248 Ngoma, Zambia. BM(NH) 70.2632 Balovale, Zambia. (Fig. 6e) BM(NH) 89.5.1.5 Caconda, Angola (Syntype of Vesperus bicolor Bocage, 1889). (Fig. 9e) Pipistrellus bodenheimeri HZM 3.3786 Jazirat al Abid, Aden, South Yemen. HZM 5.8279 Bin Gedi, Israel. (Fig. 90 Pipistrellus savii BM(NH) 31.11.11.13, BM(NH) 66.4644 E slope of Mount Olympus, Greece. BM(NH) 61.395 Ainab, Lebanon. (Fig. 6a) Pipistrellus arabicus HZM 4.10060 Wadi Sahtan, Oman. HZM 5.1 1625 Wadi Fidah, Dank/Ibri, Oman. (Fig. 7a) Pipistrellus helios BM(NH) 39.133 N Guaso Nyiro, Kenya. (Fig. 6d) BM(NH) 69.207 Kangatet, S Turkana, Kenya. HZM 2.4086 Archer's Post, Northern Frontier District, Kenya. Pipistrellus nanus BM(NH) 49.484 Kontaur, Gambia. HZM 3.2778 Sokoto, N Nigeria. HZM 3.4026, HZM 4.4027 Near Monrovia, Liberia. HZM 83.4387 Haroni-Lusitu Beacon 74, Zimbabwe. HZM 107.3212 Kabompo Boma, Zambia. HZM 146.5161, HZM 147.5162 Rondo, Lindi, Tanzania. HZM 165.5321 Liwale, Tanzania. HZM 200.6581 Karonga, Malawi. HZM 258.1 1469 Kunyale Stream, Mwinilunga District, Zambia. 274 J. E. HILL & D. L. HARRISON HZM 260.12175 Lamto, Ivory Coast. HZM 26 1 . 1 2 1 76 Ivory Coast. HZM 263.12451, HZM 264.12452 Kamuani Area, Machakos District, Kenya. (Fig. 6b, HZM 263.12451) Pipistrellus pulveratus BM(NH) 79.702 Near Nicholson Goat Bungalows, Hong Kong I. BM(NH) 79.903 Peace Mansion, Tai Hang Road, New Territories, Hong Kong (Fig. 8c) Pipistrellus Hesperus BM(NH) 98.3.1.8 Sierra Laguna, Baja California, Mexico. BM(NH) 29.1 1.7.10 Panamint Mts, California, USA. HZM 4.1 1219 Sycamore Well, Hidalgo County, New Mexico, USA. (Fig. 8d) Pipistrellus eisentrauti BM(NH) 84.1684, BM(NH) 84.1686 Mount Cameroun, Cameroun. (Fig. 9g, BM(NH) 84.1684) Pipistrellus imbricatus BM(NH) 9.1.5.286 Buitenzorg, Java. (Fig. 9a) Pipistrellus macrotis BM(NH) 23.1.2.12 Sabang, NW Sumatra. (Fig. 9b) Pipistrellus kitcheneri BM(NH) 10.4.5.47 Boentok, Barito River, Kalimantan, SC Borneo. (Fig. 8e) Pipistrellus lophurus BM(NH) 14.12.1.6 Maliwun, Victoria Province, Tenasserim, Burma (Holotype). (Fig. 8f) Pipistrellus stenopterus BM(NH) 60.1537 Institute of Medical Research Compound, Kuala Lumpur, Malaya. BM(NH) 65.135 Pasir Road, Kuala Lumpur, Malaya. (Fig. 7h) Pipistrellus (Falsistrellus) Pipistrellus affinis BM(NH) 83.3.3.2 Wynaard, India. (Fig. 8a) BM(NH) 72.4224 Argarawa, Nevrawa Elwa, Central Province, Sri Lanka. Pipistrellus peter si BM(NH) 23. 1 .2.3. Buru I, Molucca Is (Fig. 8b) Pipistrellus tasmaniensis HZM 1.8712 Barrington Tops National Park, New South Wales, Australia. (Fig. 8g) Pipistrellus (Neoromicid) Pipistrellus capensis BM(NH) 32.9.1.249 Broken Hill, Zambia. BM(NH) 54.859 Elizabethville, Zaire. BM(NH) 61.1078 Doddieburn Ranch, West Nicholson, Zimbabwe, 2300 ft, 2124'S, 2921'E. BM(NH) 72.4383 E of Lake Margharita, Bulcha Forest, Ethiopia, 1800 m, 06 1 1'N, 3610'E. BM(NH) 72.4391 Didessa River, Wollega Province, Ethiopia, 1 190 m, 0902'N, 3609'E. (Fig. 12g) BM(NH) 75.561 Mole National Park, Ghana. (Fig. 12b) BM(NH) 83.200 Mcheni Gorge, Chizarira National Park, Binga Province, Zimbabwe, 1740'S, 2752'E. HZM 36.4514 40 m NW of Serowe, Botswana. BM(NH) 66.6057 Ambositra, Madagascar (matrokd). (Fig. 12a) BM(NH) 77.2.19.6 Anzahameru, Madagascar ('minutus'). (Fig. 12i) Pipistrellus guineensis BM(NH) 70.2224, BM(NH) 70.2228, BM(NH) 72.4373 Gambela, Ethiopia, 815'N, 3435'E (BM(NH) 72.4373 at 515 m) (Fig. 12c, BM(NH) 70.2224) BM(NH) 76.293 Shagamu, Nigeria. BM(NH) 84.1019 Bontioli, Bougouriba River, Burkina Faso (Upper Volta). Pipistrellus melckorum BM(NH) 83.216 Mcheni Gorge, Chizarira National Park, Binga Province, Zimbabwe, 1740'S, 2752'E. (Fig. 12Q VESPERTILIONINE SYSTEMATICS 275 Pipistrellus somalicus BM(NH) 70.484 Mouth of Fincha River, Blue Nile Gorge, Ethiopia, 1003'N, 3720'E. (Fig. 12h) BM(NH) 76.814 S bank of Ganale Doria, Sidam-Bale Bridge, Sidamo Province, Ethiopia, 545'N, 3937'E. BM(NH) 84.1016 Comoe River, Burkina Faso (Upper Volta), 260 m, 957'N, 438'W. CMNH MJS 2846 Snai Sugar Plantation, l km S, \ km E of Giohar, Somalia, 246'N, 453 1 'E. Pipistrellus zuluensis BM(NH) 83.212 Mchesu River, Chizarira National Park, Binga Province, Zimbabwe, 1747'S, 2739'E. BM(NH) 83.215 Singama, Sibuwa, Binga Province, Zimbabwe, 17 C 36'S, 2751'E. (Fig. 12d) Pipistrellus rendalli BM(NH)89.12.12.1 Bathurst, Gambia. BM(NH) 7.12.17.1-2 Gondokoro, White Nile, Sudan. BM(NH) 23.4.12.1-2 Bugala, Sesse Is, Victoria Nyanza, Uganda. (Fig. 12e, BM(NH) 23.4.12.2) BM(NH) 48.702 N'ko, Obubra Division, S Nigeria (Ibrunneus). (Fig. 14b) Pipistrellus tenuipinnis BM(NH) 47.350 Umuahia, E Nigeria. BM(NH) 54.917 Bonthe, Sierra Leone. BM(NH) 67.1734 Bota, Victoria, Cameroun, 400'N, 905'E. (Fig. 12j) Pipistrellus (Arielulus) Pipistrellus circumdatus BM(NH) 73.618 Telecommunications Tower, Fraser's Hill, Pahang, Malaya. (Fig. 2e) Pipistrellus cuprosus BM(NH) 83.351 Sepilok, Sabah, Borneo, 552'N, 1 1756'E (Holotype). (Fig. 9h) Pipistrellus societatis BM(NH) 67.1605 Base Camp, Gunong Benom, Pahang, Malaya, 800 ft (Holotype). (Fig. 9c) Nyctalus noctula BM(NH) - - Locality unknown. HZM 10.613 Bottisham, Cambridgeshire, England. HZM 33.8888 Winchelsea Beach, Sussex, England. (Fig. 100 Laephotis botswanae BM(NH) - - Zomba, Malawi (original No. 2269; damaged). Laephotis wintoni HZM 1.3020 Nyeri, Mount Kenya, Kenya. (Fig. 160 Glischropus tylopus BM(NH) 10.4.5.136 Upper Barito River, Kalimantan, SC Borneo. (Fig. 18a) Scotozous dormeri BM(NH) 12.3.8.30 Furdapur, Ajanta, Khandesh, India. BM(NH) - - Kathiawar, India (Original No. BNHS 2007). (Fig. 1 6d) Scoteanax rueppellii BM(NH) 80.3.25.1 Richmond River, New South Wales, Australia. (Fig. 16i) Scotorepens balstoni BM(NH) 10.6.21.9 Hermannsburg, Northern Territory, Australia. (Fig. 16g) Scotorepens greyii BM(NH) 75.2261 Pine Creek, 20 m ESE of Candy's Hill, Northern Territory, Australia, 1349'S, 13149'E. (Fig. 16h) Nycticeinops schlieffenii BM(NH) 14.7.3 1.14 Wei Wei River, Kenya. BM(NH) 15.3.6.66 Kamisu, Dinda River, Sudan. BM(NH) 71.675 Awash, Filhoa, Ethiopia, 0900'N, 3858'E. HZM 5.2120 Ikau, Rukwa, Tanzania. (Fig. 16e) Scotoecus albigula BM(NH) 63.1042 Calundo, Lunda, Angola. (Fig. 20a) 276 J. E. HILL & D. L. HARRISON Scotoecus albofuscus BM(NH) 96. 1 2.3 1 . 1 Sierra Leone. (Fig. 20e) Scotoecus hindei BM(NH) 14.7.31.13 30m NW of Baringo, Kenya. (Fig. 20d) BM(NH) 66.1466 Jos, Nigeria (falabae). (Fig. 20b) Scotoecus hirundo BM(NH) 76.771 Mole National Park, Ghana. (Fig. 20c) Scotoecus pallidus BM(NH) 86.531 Afghanistan (damaged). Philetor brachypterus BM(NH) - - New Guinea. (Fig. 16b) Hesperoptenus (Milithronycteris) Hesperoptenus blanfordi BM(NH) 83.853 Sepilok, Sabah, Borneo, 552'N, 1 1756'E. (Fig. 21g) Hesperoptenus tickelli BM(NH) 71.12.26.1 Sri Lanka. (Fig. 21b) Hesperoptenus tomesi BM(NH) 7.1.1.428 Malacca, Malaya (Holotype). (Fig. 2 la) Chalinolobus gouldi BM(NH) 71.1504 Westwood, near Rockhampton, Queensland, Australia. (Fig. 17b) Chalinolobus morio BM(NH) 6.8.1.60 (King River, Western Australia. (Fig. 17a) Chalinolobus nigrogriseus BM(NH) 44.6.13.2 Port Essington, Northern Territory, Australia (rogersi). BM(NH) 75.2260 Pine Creek, 20 m ESE of Gandy's Hill, Northern Territory, Australia, 1349'S, 13149'E. (rogersi). (Fig. 17c) Chalinolobus picatus BM(NH) 9.3.7.2 Gunnamulla, Queensland, Australia. (Fig. 17d) Chalinolobus tuberculatus BM(NH) 89.10.27.1 Outlying islands near Stewart I, New Zealand. (Fig. 17e) Nyctophilinae Nyctophilus bifax BM(NH) 67.5.6.5 Cape York, Queensland, Australia. BM(NH) 77.3.28.1 Islands of Torres Straits, Australia. BM(NH) 86.1 1.8.12 Somerset, Cape York, Queensland, Australia. BM(NH) 15.3.13.1 Cloncurry, Queensland, Australia. BM(NH) 15.3.13.3 Herberton District, Queensland, Australia (Holotype). (Fig. 22a) Nyctophilus daedalus BM(NH) 47.7.2 1 . 1 6, BM(NH) - - Port Essington, Northern Territory, Australia. (Fig. 22g, BM(NH) 47.7.21.16) BM(NH) 97.4.12.5 Daly River, Northern Territory, Australia. Nyctophilus gouldi BM(NH) 15.3.13.7 Ash I, Hunter River, New South Wales, Australia (damaged, part lost). BM(NH) 15.3.13.8 Sydney, New South Wales, Australia. BM(NH) - - Botany, Sydney, New South Wales, Australia (Original No. 164) (Fig. 22d) HZM 1.12085 Werrikimbe, Hastingsshire, New South Wales, Australia. (Fig. 16c) BM(NH) 52. 1 . 1 5.30 Tasmania (sherrini). (Fig. 220 Nyctophilus geoffroyi BM(NH) 15.3.13.1 1 Kosciusko, New South Wales, Australia (pacificus). BM(NH) - - Tasmania (pacificus) (Original No. M.I 735). VESPERTILIONINE SYSTEMATICS 277 BM(NH) - - Launceston, Tasmania (pacificus) (Original No. M. 168) (Fig. 22e) BM(NH) 7.1.4.3 Alexandria, Northern Territory, Australia (pallescens). (Fig. 22b) Nyctophilus microtis BM(NH) 88.4.18.1 Sogeri, Papua New Guinea (Holotype). (Fig. 22c) Pharotis imogene BM(NH) 97.8.7.21 Kamali, Papua New Guinea. (Fig. 22h) 278 J. E. HILL & D. L. HARRISON VESPERTILIONINE SYSTEMATICS 279 280 J. E. HILL & D. L. HARRISON Table 2 Usual incisive and premolar dental formulae in the Vespertilioninae and Nyctophilinae. Total number of teeth (including four canines and twelve molars) in parentheses. Dental notation of Miller ( 1 907). (38) Myotis, Pizonyx (36) Lasionycteris, Plecotus, Idionycteris, Eudiscopus (34) Euderma, Barbastella, la, Pipistrellus, Glischropus, Scotozous, Nyctalus, Chalinolobus (32) Eptesicus, Vespertilio, Histiotus, Tylonycteris , Mimetillus, Glauconycteris, Pipistrellus, Laephotis, Philetor, Hesperoptenus (32) Lasiurus (30) Rhogeessa, Baeodon, Nycticeius, Otonycteris, Dasypterus, Scotomanes, Scotophilus, Scoteanax, Scotorepens, Nycticeinops, Scotoecus, Nyctophilus, Pharotis (28) Antrozous, Bauerus Table 3 Classifications of the Vespertilioninae and Nyctophilinae. That of Tate (\942a) is concerned primarily with Oriental and Australasian taxa, those of Koopman with Australasian (1973) and predominantly African (1975) forms. Tate(1942a) Koopman (1973, 1975) Hill & Harrison Pipistrellus abramus group abramus akokomuli bancanus camortae irretitus paterculus pumiloides pipistrellus group pipistrellus (Including bactrianus) nathusii coromandra group aladdin angulatus collinus coromandra imbricatus meyeni micropus murrayi ponceleti portensis regulus sturdeei subulidens tramatus tenuis group mimus (Including glaucillus) nitidus papuanus (Including orientalis) principulus tenuis ceylonicus group ceylonicus (Including chrysothrix, indicus, subcanus) Pipistrellus Amalgamates pipistrellus, abramus ( =javanicus), coromandra and tenuis groups of Tate (1942a) pipistrellus group imbricatus javanicus (Including abramus) meyeni nanus (Including (?) Helios) permixtus tenuis (Including angulatus, collinus, nitidus, papuanus, ponceleti, murrayi, sewelanus, subulidens, westralis [Koopman, 1984c]) ceylonicus group ceylonicus Pipistrellus Pipistrellus (Pipistrellus) pipistrellus group pipistrellus subgroup pipistrellus (Including aladdin, bactrianus, lacteus, mediterraneus) nathusii permixtus javanicus subgroup abramus (Including akokomuli, irretitus, pumiloides) babu endoi javanicus (Including bancanus, camortae, meyeni, 'tralatitius") paterculus peguensis coromandra subgroup adamsi angulatus (Including ponceleti) collinus coromandra (Including afghanus, portensis, tramatus) mimus (Including glaucillus, principulus) murrayi papuanus sturdeei tenuis (Including nitidus, sewelanus, subulidens) wattsi westralis ceylonicus subgroup ceylonicus (Including borneanus, chrysothrix, indicus, raptor, shanorum, subcanus) (?) minahassae VESPERTILIONINE SYSTEMATICS 281 Table 3-cont. Tate(1942a) Koopman (1973, 1975) Hill & Harrison minahassae group minahassae rueppellii group coxi kuhlii group babu canus kuhlii (Including ikhwanius, lepidus) leucotis lobatus Eptesicus pumilus group pumilus (Including caurinus, darlingtoni, vulturnus) pygmaeus Pipistrellus savii group austenianus cadornae curtatus macrotis savii vordermanni minahassae group minahassae rueppellii group rueppellii (Including (?) fuscipes; pulcher) kuhlii group aero anchietae deserti inexspectatus kuhlii (Including (?) aegyptius; fuscatus) rusticus (Including marrensis) savii group ariel macrotis maderensis Hesperus group hesperus musciculus joffrei group anthonyi joffrei stenopterus joffrei group stenopterus rueppellii group crassulus nanulus rueppellii (Including coxi, fuscipes, leucomelas, pulcher, senegalensis , vernayi) kuhlii group aero deserti inexspectatus kuhlii (Including (?) aegyptius; fuscatus, ikhwanius) maderensis rusticus (Including marrensis) Pipistrellus ( Vespadelus) douglasorum pumilus (Including darlingtoni) regulus sagittula vulturnus Pipistrellus (Perimyotis) subflavus Pipistrellus (Hypsugo) savii group savii subgroup anchietae ( = 'bicolor'7) ariel austenianus bodenheimeri savii (Including caucasicus, darwini, maurus) nanus subgroup arabicus helios musciculus nanus (Including culex, stampflii) pulveratus subgroup pulveratus hesperus subgroup hesperus eisentrauti subgroup eisentrauti imbricatus subgroup curtatus imbricatus macrotis vordermanni lophurus subgroup cadornae kitcheneri lophurus stenopterus group anthonyi joffrei stenopterus ^.......:n.<^ v ,,.:... ; :... 1 ......,.,. " w " r '^ > Fig. 5 Baculum (D, RL) of a, Pipistrellus kuhlii; b, P. maderensis; c, P. deserti; d, P. rusticus. Scale = 0-5 mm. VESPERTILIONINE SYSTEMATICS 289 Fig. 6 Baculum of a, Pipistrellus savii (D, RL); b, P. HO/IMS (D, RL); c, P. rusticus (D, RL); d, P. helios (D, RL); e, P. anchietae (D, LVL, reversed). Scale = 0-5 mm. S?>t^KW^^ .,,,...-,.,.:;:.,. , : -..,.. wmc.r.l.UW Fig. 7 Baculum of a, Pipistrellus arabicus (D, RL, RVL); b, P. coromandra (tramatus) (D, RL); c, P. coromandra (D, RL); d, P. ceylonicus (D, RL); e, P. crassulus (D); f, P. nanulus (D, RL); g, P. mimus (D, RL); h, P. stenopterus (D, RL, RVL). Scale = 1 mm. Fig. 8 Baculum of a, Pipistrellus affinis (D, RL); b, P. peter si (D, RL); c, P. pulveratus (D, RL, RVL); d, P. Hesperus (D, LL, reversed, LVL); e, P. kitcheneri (D, RL, RVL); f, P. lophurus (D, RL); g, P. tasmaniensis (D, RL, V). Scale = 1 mm. Fig. 9 Baculum (D, RL except where stated) of a, Pipistrellus imbricatus; b, P. macrotis; c, P. societatis, d, P. tennis (nitidus) (D, RL, RVL); e, P. anchietae ('Vesperus' bicolor); f, P. bodenheimeri; g, P. eisentrauti; h, P. cuprosus. Scales a-g= 1 mm; h = 0-5 mm. _l Fig. 10 Baculum (D, RL) of a, Pipistrellus rueppellii (pulcher); b, P. rueppellii; c, P. adamsi; d, P. westralis; e, P. javanicus; f, Nyctalus noctula; g, P. wattsi; h, P. mackenziei (c, d, g, h from Kitchener etal, 1986). Scales = a, b, e, f=2mm;c, d, g, h= 1 mm. 294 J. E. HILL & D. L. HARRISON Fig. 1 1 Baculum (V, RL) of a, Pipistrellus pumilus pumilus; b, P. pumilus (caurinus); c, P. vulturnus; d, P. douglasorum; e, P. regulus; f, P. sagittula (a-c, e, f from McKean et al., 1970; d from Kitchener, 1976). Scale = 2 mm. Fig. 12 Baculum (D, RL) of a, Pipistrellus capensis (matrokd); b, P. capensis; c, P. guineensis; d, P. zuluensis; e, P. rendalli (with anterior view); f, P. melckorum; g, P. capensis; h, P. somalicus', i, P. capensis ('minutus'); j, P. tenuipinnis; k, P.pumilus. Scale = 1 mm. Fig. 13 Baculum (D, RL) of a, Eptesicus fuscus; b, E. hottentotus (megalurus); c, E. furinalis; d, E. brasiliensis (andinus); e, E. bobrinskoi; f, E. floweri; g, E. serotinus; h, E. serotinus (isabellinus); i, E. fuscus (hispaniolae); j, . bottae (innesi); k, . brasiliensis: 1, E. flower i(lowei). Scale = 1 mm. VESPERTILIONINE SYSTEMATICS 297 Fig. 14 Baculum (D, RL except where stated) of a, Eptesicus bottae (omanensis); b, Pipistrellus rendalli (? brunneus); c, Eptesicus nasutus; d, Plecotus teneriffae (D) (from Ibanez & Fernandez, 1986). Scales = 1 mm. d %**^ I I Fig. 21 Baculum (D, RL except where stated) of a, Hesperoptenus tomesi, b, H. tickelli, c, H. doriae (a-c from Hill, 1 976); d, la io (D) (from Topal, 1 970); e, Scotorepens orion, f, S. sanborni (e, f from Kitchener & Caputi, 1985); g, Hesperoptenus blanfordi (from Hill & Francis, 1984); h, Scotoecus pallidus (from Agrawal & Sinha, 1973); i, Vespertilio murinus (V, RL) (from Topal, 1958); j, V. orientalis (from Wallin, 1969). Scales a-c = 2 mm; d-h, j= 1 mm; i = 0-5 mm. _- ' - Fig. 22 Baculum (D, RL) of a, Nyctophilus bifax; b, AT. geoffroyi (pallescens); c, M microtis; d, Af. gouldi; e, A 1 , geoffroyi (pacificus); f, A 7 ^. gouldi (sherrini); g, A 7 , daedalus; h, Pharotis imogene. Scale = 2 mm. Hill, J. E. and Harrison, David L. 1987. "The baculum in the Vespertilioninae (Chiroptera: Vespertilionidae) with a systematic review, a synopsis of Pipistrellus and Eptesicus, and the descriptions of a new genus and subgenus." Bulletin of the British Museum (Natural History) Zoology 52, 225–305. https://doi.org/10.5962/p.18307. View This Item Online: https://www.biodiversitylibrary.org/item/19520 DOI: https://doi.org/10.5962/p.18307 Permalink: https://www.biodiversitylibrary.org/partpdf/18307 Holding Institution Natural History Museum Library, London Sponsored by Natural History Museum Library, London Copyright & Reuse Copyright Status: In copyright. Digitized with the permission of the rights holder. Rights Holder: The Trustees of the Natural History Museum, London License: http://creativecommons.org/licenses/by-nc-sa/4.0/ Rights: http://biodiversitylibrary.org/permissions This document was created from content at the Biodiversity Heritage Library, the world's largest open access digital library for biodiversity literature and archives. Visit BHL at https://www.biodiversitylibrary.org. This file was generated 9 October 2023 at 18:56 UTC