Nutlet production and germination in female and hermaphrodite plants of Thymus albicans Hoffmanns. & Link (Lamiaceae)

Benito Valdés, Zoila Diaz Lifante & Raquel Parra Nutlet production and gerrnination in fernale and herrnaphrodite plants of Thymus albicans Hoffrnanns. & Link (Lamiaceae) Introduction Abstract Valdés, 8., Diaz Lifante, Z. & Parra, R.: Nutlet production and gerrnination in female and hermaphrodite plants of Thymus albicans Hoffmanns. & Link (Lamiaceae ). -Boeeonea 13: 353360.2001. -ISSN 1120-4060. Some reproductive aspects are studied in Thymus albicans Hoffmanns. & Link, endemie to SW Iberian Peninsula. Pereentage of herrnaphrodite plants fruit set, nutlet production, seed set, reproductive effieiency and pe rc entage of gerrnination of seeds produced by female and hermaphrodite plants are studied. Percentage of female and herrnaphrodite plants is very variable be tween populations. Nutlet produetion and reproduetive efficiency a re higher in female than in herrnaphrodite plant s, and this, together with a slightly advaneed gerrnination of seed pro-. duced by females, may inerease female fitness in natural populations. Gynodioecy is the occurrence offemale and herrnaphrodite individuals in the same natural populations of a species. It is frequently the result of the regular occurrence of male sterile genes in otherwise herrnaphrodite populations (Richards 1986). Female plants can be considered as herrnaphrodites that have lost their male function, a process in which both nuclear and cytoplasmic genes may be involved (Belhassen & al. 1991). Gynodioecy can be considered a well-stablished breeding system in the Lamiaceae; within a population, frequencies of herrnaphrodite and female plants show little variation with time (Richards 1986). Models of gynodioecy involving the effects of selfing postulate that female are maintained in the population because their offspring are obligatory outcrossed and hence more fit than the progeny of herrnaphrodites often resulting from selffertilization (Cox 1988). Gynodioecy occurs in most species of Thymus (Morales 1986) and has been studied in detail in T vulgaris L. by Dommee & al. (1978) and Assouad & al. (1978). As in many other gynodioecious Angiosperrns (Delph 1996), the corolla in ~male flowers is smaller than in herrnaphrodite flowers, and while in the latter there is a well marked protandry, with the stigma becoming receptive four days after anthesis, in female flowers the stigma is receptive inmediately after flower opening (Morales 1986). Thymus albicans Hoffrnanns.\& Link is one ofthe two Iberian Peninsula species of sect. 354 Valdés & al.: Nutlet production and germination in Thymus Mastichina (Mi11er) Bentham, characterized by their flat leaves, which are never ciliate at the base, and capituliform flower-heads with small white flowers with ciliate calyx-teeth. T albicans differs from the other species in the section, T mastichina (L.) L. , by the size of the calyx, higher in T albicans, and size of calyx teeth, as long as or shorter than tube in T mastichina, but quite longer than the tube in T albicans. It is a gynodioecious endangered species endemie of SW Iberian Peninsula where it lives in coastal sandy soils of a reduced area in S Portugal around Faro (Algarve) and in Cadiz province (Spain), where its distribution area is now in regresion. Hs presence in Huelva and Sevi11a provinces (Morales 1986) has not been recently confirmed. The aims of this paper is to quantify the sexual reproductive capacity of this species. Sex distribution, nutlet production, reproductive efficiency and seed germination capacity have been studied in several populations. Possible differences between female and hermaphrodite plants for these parameters have been checked. Materia) and Methods Five populations were chosen near Chiclana (Oidiz province) (Fig. 1), a11 living on sandy soils with similar climatic conditions. Populations l, 2 and 3 grow as part of the Fig. l. Generai distribution of Thymus albicans Hoffrnanns. & Link. and origin of the populations studied. Bocconea 13 -2001 355 open underscrub of Pinus pinea forest, and populations 4 and 5 in open and sunny areas around the forest. Percentage of female and hermaprodite plants was estimated by counting ali specimens in four populations. Fruit set (% frlfl), nutlet production, seed set (% S/O) and percentage of abortion was measured in several female and hermaprodite plants in the four populations in 1995 and 1996. Plants of similar size, separate for an equivalent distance and distributed at random in each population were chosen. Five replicates of 50 calyces per ('lant, ali separate at random, were used to evaluate these parameters. Fruit set was estimated as percentage of flowers which produced fruiting calyces, this is, with at least one nutlet. Nutlet production per calyx was estimated as number ofnutlets produced by each fruiting calyx. Seed set was estimated as percentage of ovules which produced nutlets. Percentage of abortion refers to ali fruiting calyces studi ed; aborted fertilized ovules were recognizable as seedless nutlets, which were always smaller than ripe ones. Reproductive efficiency was measured according to Wiens & al. (1987) as % S/O x frlfl. Percentage and dynamic of germination of seeds produced by female and hermaphrodite plants of four populations were studi ed in laboratory conditions. Four replicates of each sample, of25 nutlets each, were sown in October 1995 and December 1996, in sterile Petri dishes with wet filter paper. Seeds were grown in natural day light with temperatures ranging from 18 to 28°C. Seed germination was followed daily during 40 days. The number of days elapsed from sowing to germination ofthe first seed (tO) and from sowing to 50% germination (t50) have been used as indicators of germinaton dynamic. Results Percentages oJ gynodioecy. -Size of population, number of plants, density and sex distribution in the five populations studi ed are indicated in table 1. Percentage of hermaphrodite and female plants is very variable from one population to another, ranging from 39.13% to 63.15% ofhermaphrodites. In these perennial plants, no differences in sex percentages have been observed in two consecutive years (1995 and 1996) except for population l, which was partially destroyed by building activities in the spring of 1996. Fructification.- Fruit set, nutlet production, seed set, reproductive efficiency (PERS) and percentage of abortion in female and hermaphrodite plants of the four populations Table l. Density and sex distribution in the population studi ed. Population Surface (m2) N° plants N° plants/m2 % Hermaphrodite 360 61 0.16 46. 15 2 675 83 0.12 39.13 3 372 95 0. 25 40.47 4 42 19 0.45 63 . 15 5 4234 2032 0.47 62.22 356 Valdés & al.: Nutlet production and gennination in Thymus Table 2. Fructification in 1995 and 1996 in female (F) and hermaphrodite (H) plants of the populations 1,3,4 and 5. N, number ofplants studied . A) 1995 Population N % fr / fl nutlets/ cal yx %S /O PERS % (mean ± s.d.) abortion 3-F 3 }' 0.39 4.20 0.38 9.71 1.01 57.63 3-H 4 12.41 7.29 0.60 15.25 1.89 40.46 4-F 4 57.19 16.59 0.65 16.29 9.31 40.68 4-H l 11.36 4.10 0.41 10.40 1.18 58.32 5-F 5 29.34 13.42 0.68 17.24 5.06 38.70 5-H 4 13 . 61 8.13 0.29 7.08 0.96 70.68 B) 1996 Population N % frlfl nutlets/calyx %S/O PERS % (mean ± s.d.) abortion l - F 4 17.62 5.55 0.48 11.98 2.1/ 52.06 l - H 4 10.15 6.86 0.42 10 . 58 1.07 41.03 3-F 9 36.68 10.67 0.64 15.91 5.84 34.60 3 -H 9 20.24 9.58 0.37 9.28 1.88 54.90 5-F 4 14.28 1.42 0.39 9.75 1.39 60.99 5 - H 4 20.78 10.34 0.39 9.82 2.04 54.47 studi ed in 1995 and 1996 are indicated in table 2. Percentages of fruiting calyces with O, l and 2 ripe nutlets in female and hennaphrodite plants are shown in Figs. 2a and 2b. No statistical tests have been carried out owed to the heterogeneity in the number of plants studied in each population. In populations l and 4 nutlet production and reproductive efficiency are higher in female than ÌÌì'~ennap hrodite plants. In populations 3 and 5, where these values were estimated for two consecutive years, results are rather contradictory. In population 3 both nutlet production per fruiting calyx and fruit set were 10wer in female than in hennaphrodite plants in 1995, but considerably higher in 1996. In population 5, on the contrary, both parameters were quite 'higher in female than in hennaphrodite plants in 1995, while they were very similar in 1996. As shown in Figs, 2a and 2b, the number of ripe nutlets per calyx is rather low, with a reduced percentage of calyces with two ripe nutlets. No calyx was found with three or four Bocconea 13 -2001 357 ripe nutlets. Number of nutlets produced by ripe calyx tends to. be higher in female than in hermaprodite plants. Seed germination.- Percentage and dynamics of germination of seeds for four populations studi ed and level of statistic significant differences for Iukey HSD test are given in table 3. Ihere are no statistically significant differences between populations 3 and 4 as studi ed in 1995. In 1996 only population 1 showed a percentage of seed germination sig- % % 10 8 6 4 2 10 O 8 O 6 O 4 O 2 O 3-F 3-H 1-F 1-H a) 4-F 4-H b) 3-F 3-H 5-F 5-F D O nut ~ 1 nut .2 nut 5-H D Onut ~ 1 nut .2 nut 5-H Fig. 2. Percentage of fruiting calyces with 0, l and 2 nutlets in female (F) and hermaphrodite (H) plants: a, data for populations 3, 4 and 5 (1995); b, data for populations I, 3 and 5 (1996). 358 Valdés & al.: Nutlet production and germination in Thymus nificantly lower than in populations 3 and 5, which did not differ one from the other. No differences have been found either between 1995 and 1996 in population 2. Data for female and hermaphrodite plants of these populations are shown separately in table 3. The statisti c comparison between pairs of means in female versus hermaphrodite plants (Mann-Whitney-U non parametri c test) has shown only significantly differences at the 0.05 level in population 3 for 1995, but at the 0.1 level in population 5 for 1996, with a lower percentage in female versus hermaphrodite plants. Data on dynamics of germination of seeds are also given in table 3, where the parameters tO and t50 have been indicated. Germination usually begins earlier in nutlets produced by female than by hermaphrodite plants, but in populations 3 no difference was found in 1996. The parameter t50 is ussually reached 1-3 days earlier in nutlets produced by females than in those produced by hermaphrodites, but once more population 3 differs at this respect. Table 3. Percentage (%) and dynamics (tO and t50, in days) of germination ofseeds in female (F) and hermaphrodite (H) plants from populations 1,3, 4 and 5. Mean and standard deviation (in brackets) for % germination, and mean and range of variation (in brackets) for tO and t50, are indicated. The same letter upper means implies statistic differences with significance at leve I as follows: a= p< O.OI, b= p< O.OOI [Tukey HSD test]; x= p< 0.05, y= p<O.1 [Mann-Whitney-U]. Population year sex % tO t50 nO replico l 1996 F 49.0 (8.9) 3 (2 -5) 7 (4-8) 4 H 44.0 (-) 4 (4) 5 (5) l Total 48.0 a,b (0.8) 5 3 1995 F 59.0 x (22.2) 3 (3) 4 (4) 4 H 84.0 x (3.3) 3.5 (3-4) 7 (7) 4 Total 71.5 (20.0) 8 1996 F 81.0 (7.6) 3 (3) 4.5 (4-6) 4 H 85.0 (8.2) 3 (3) 4 .2 5 (3-5) 4 Total 82.8 a (7.6) 8 4 1995 F 83.0 (8.9) 3.5 (3-4) 6.2(4-7) 4 H 68 .6 (14. 1) 4 (4) 10.5 (7-14) 2 Total 78.8 (11.6) 6 5 1996 F 78 .0 y (4.0) 2.5 (2-3) 3 (3) 4 H 86.0 y (6.9) 2 (2) 4 (3 -5) 4 Total 82.0 b (16.4) 8 Bocconea 13 - 2001 359 Discussion There is a wide variability in the fructification parameters analyzed, which together with the differences in number of individua1s studi ed from each popu1ation comp1icates statistical comparisons. However, data suggest that percentage of fructification is very low in ali populations. Percentage of autogamy in hermaphrodite plants is equally very low, as it has been observed a percentage of fructification lower than 5%, with a PERS of c. 0.1 (unpublished data). This indicates the low survival potenciality of this species. Seed abortion, sometimes as high as 70%, seems to greatly affect fruit set, and there is a negative correlation between number of nutlets per calyx and degree of seed abortion, as deduced from table 2. Besides, a high, but not evaluated, style predation by smalliarvae has been observed in most female and hermaphrodite plants. Bothseed abortion and style predation could explain the low fruit set observed, and also the differences found in a particular population in consecutive years. On the other hand the results show a generai tendency to a higher nutlet production and reproductive efficiency in female than in hermaphrodite plants, which agree with the generai tendency found in gynodioecious populations of other species of Thymus (Dommèe & al. 1978, Assouad & al. 1978), and with results found by several authors in other groups (Maki 1993, Eckhart 1992). Seed germination percentage is generally high, with a total mean of c. 74%. This indicates that there are not special requirements for seed germination and that this does not make any problem for the survival of this species. Population l is the only one which significantly differs from the others by its low germination percentage, but no logical explanation has been found to this difference, as population l is not the smallest and occupies a similar habitat that the other three. At the other side, as deduced from table 3, there is no relationship between percentage of germination and sexuality. Only significant differences are found in population 3 in 1995, but not in the same population in 1996. As indicated above, nutlets produced by female plants seem to germinate earlier than those produce by hermaphrodites. As germination occurs after Autumn rainfall, this difference in germination may favour female offspring. However, further observations could confirm this tendency. References Assouad, M.W., Dommèe, B., Lumaret, R. & Valdeyron, G. 1978: Reproductive capaciti es in the sexual forms ofthe gynodioecious species Thymus vulgaris L.-Bot. J. Linn. Soc. 77: 29-39. Belhassen, E., Dommèe, B., Atlan, A., Gouyon, P. H ., Pomente, D., Assouad, M. W. & Couvet, D. 1991: Complex determination ofmale sterility in Thymus vulgaris L.: genetic and molecular analysis. -Theor. Appl. Genet. 82: 137-143. Cox, P. A. 1988: Monomorphic and dimorphic sexual strategies: a modular approach.- Pp. 80-97 in: Doust, J. L. (ed.), Plant reproductive ecology. Pattems and strategies. -Oxford. Delph, L. F. 1996: Flowersize dimorphism in plants with unisexual flowers. -Pp. 217-237 in: Lloyd, D. G. & S. C. H. Barret (eds.), Fiorai biology. -New York. Dommèe, B., Assouad, M. W. & Valdeyron, G: 1978: Natural selection and gynodioecy in Thymus vulgaris L. -Bot. J. Linn. Soc. 77: 17-28. Eckhart, V. M. 1992: Resource compensation and the evolution of gynodioecy in Phacelia linearis (Hydrophyllaceae). -Evolution 46: 1313-1328. 360 Valdés & al.: Nutlet production and germination in Thymus Maki, Y. 1993: Outcrossing and fecundity advantage of females in ginodioecious Chionographis japonica varo kurohimensis (Liliaceae). -Am. J. Bot. 80: 629-634. Morales, R. 1986: Taxonomia de los géneros Thymus (exc\uida la secci6n Serpyllum) y Th ymbra en la Peninsula Ibérica .. -Ruizia 3: 1-324. Richards, A. J. 1986: Plant breeding systems. -London. Wiens, D., Calvin, C. L., Wilson, C. A., Darven, C. L, Frank, D. & Seavey, S. K. 1987: Reproductive success, spontaneous abortion, and genetic load in flowering plants. - Oecologia (Berl.) 71 : 501-782. Address of the authors: Valdés, B., Dia z Lifante, Z. & Parra R.: Dpto. Biologia Vegetai y Ecologia, Facultad de Biologia, Avda. Reina Mercedes sin. Apdo. correos 1095.41080 Sevilla, Spain.