Jun 16, 2010 - o f all pairs) had fewer long copulations, but a higher overall copulatory activity. Copulation rate is .... (rather than the last frame in copula).
Ethology 102, 2 3 6 2 5 1 (1996) 0 1996 Blackwell Wissenschafts-Verlag, Berlin ISSN 0179.1613
Max-P/anck-lnstif~t~ur l~erba/ten~bysioLogie, Seewiesen
Mating Behaviour in a Hermaphroditic Flatworm with Reciprocal Insemination: Do They Assess Their Mates during Copulation? ANNEPi:ri:Rs, ANIXEA STRFM;& NI(:OI.AAS K. Micti1r:i-s P t . n - ~ >A., , STRFW;,A. & MI( I I I I . I S , N. K. 1996: Mating hehaviour in a hermaphrodiuc flatworm with reciprocal insemination: d o they asseqs their mates during copulation? Etholoky 102, 236-251.
Abstract In the freshwater plananan fhg& po/ychmu (Tricladida, Paludicola), both animals fulfil the male and female role simultaneously in any given copulation. T h i s study presents the tirst detailed account of the copulatory behaviour, timing and frequency in this species. U ’ e also describe an experimental set-up that enables continuous, undisturhed ohsenration of large numbers o f animals. I). po/yrbma pairs copulate repeatedly in the lab (up t o tight times in 5 d). Animals that were kepr in isolation for longer, were more likely t o copulate. The mating behaviour lacks a behaviourally rccopimhle precopulatory courtship sequence, supgating that precopulaton assessment is absent o r only marginally important. Copulation duration ranged from a few minutes t o 2.5 h and showed a distinct Iimodal distribution. Two copulation hpes were recognized: short (< 35 min) and long ( 235 min). Histological analysis showed that sperm transfer is rare during short copulations, hut after long copulations all animals had received sperm and sperm transfer was reciprocal in all pairs where both partners could he investigated. Hence, only long copula ti"^^ are ‘true’ copulations with sperm transfer. Pairs with short copularims (40 ‘’4, o f all pairs) had fewer long copulations, but a higher overall copulatory activity. Copulation rate is constant over 5 d for both copulation types. I t does, however, fluctuate with the time of day: most (land copulations take place at night. Pairs that copulated more often, also produced more cocoons. We a r p e r.iat in-copula assessment during the first 30 min of the copulation determines whether copulations are sometimes interrupted before sperm are transferred and that this explains the occurrence o f short and long copulations. Corresponding author: Nicolaas K. MICHIFIS, hlax-l’lanck-lnstitut fur Verhaltensphysiologie, D82.31 9 Seewiesen, Post Starnberg, ( 5 60.6 2 2.9 min; Fig. 4), ranging from 5 min to 2.5 h with a pronounced minimum around 30-40 rnin. As a result, the distribution diverges significantly from a normal or uniform distribution (Fig. 4). Its low kurtosis (- 0.840) indicates a very platykurtic distribution, typical for a bimodal distribution (Soh;,~. & ROHLF 1981). Suspecting a functional explanation for this
243
Mating Behaviour in a Hermaphroditic Flatworm c
c
20
s 3
5
15
10
5
0
20
10
$0
60
100
120
140
Copulation duration (min)
Fli. 4; Frequency distribution of the duration of 137 copulations observed in experiments with 47 copulating pairs of L>gResiapu&hmu. The x-axis labels indicate the lower limit for each 10-min interval. The distribution differs significantly from a normal and uniform distribution (I(o1mogorov-Smirnov Z = 1.GO, p = 0.011 and Z = 3.4, p < 0.0001). A minimum in the number of copulations appears around a copulation duration of 35 min
observation, we performed all following analyses separately for ‘short’ ( 35 min) and short ( < 3.5 min) copulations for each o t 47 pairs with at least one copulation. When more than one pair had the same x and y values, the numbcr o f pairs is indicated. Pairs with more short copulations had fewer long copulations. R. is a Spearnian rank correlauon coefficient
‘short’ copulations were briefer than 17 min. The ‘long’ copulations were all longer than 44 min. Occasional obsewations o f pairs copulating with their ventral side against the clear acrylic lid of the observation chamber allowed us to see the extruded penises. From these observations, it appeared that there is intromission in both copulation types (A. PIT~ERS, pers. obs.). Occurrence of Short versus Long Copulations
Not all pairs engaged in short copulations: They were observed in only 19 (40 (Yo) o f 47 pairs. It appeared that pairs with more long copulations, had fewer short 7bMr 2: 1:ffect of the occurrence of short copulations on the number and duration of long copulati~~ns. Values are R f SE per pair o f 47 pairs o f I). po/ydmu obsened over a 5-dperiod (data set A). Six pairs had short copulations only and were excluded in the comparison o f the average durauon o f long copulations. The p values are from Xlann-Yihitncy 1’-tests O n l y long copulations
No. pairs No. copulations per pair No. long copulations ( > 3 35 min) per p a r Avg. duration o f long copulations per pair (min)
2x
2.502 0.23 2.50 2 0.21 80.0 t 2.5
At
least one short copulation
p value
19 3.53 2 0.41 1.37 2 0.34 77.8 t 0.5
O.O?O
0.002 0.34
245
Maung Behaviour in a Hermaphrodmc Flatworm
I ._
25
Pairs with long copulalions only
m
Q I
g 2.0 fn C
0 ._ w
L?!
3 Q
1.5
0
0
g
10
a,
> ._
I
m
7 05
E,
0 00
v
, 0
.~ ~
~
., .... 20
long copulations short copulations
40
60
~ 80
100
120
Time since start observations (h) f,&. 6: Cumulative number of copulations, averaged per pair during a 5-d observation penod. Filled symbols are from 28 pairs that engaged in long copulations only. Open symbols are from 19 pairs that had at least one short copulation (separate symbols for short and long copulations). The linear reqession lines are only shown to illustrate that, except for the initial 10 h, copulations occur roughly at a constant rate in all p ) u p s . The wavy pattern is a result of the diurnal copulation pattern (see Fig. 7)
copulations (Fig. 5). Table 2 confirms that, although they copulated more often in general, pairs with at least one short copulation had fewer long copulations than pairs that had long copulations only. The average duration of long copulations did not differ between both g o u p s Fable 2). Fig. 6 shows that pairs that engaged in at least one short copulation had short and long copulations at a constant rate. This indicates that there was no tendency for short copulations to occur before or after long copulations ah this would result in two non-linear curves. To look for sequence effects, we performed some additional analyses in those pairs that had the two copulation types (n = 13). The mean time elapsed since the start of the observation for the first short and the first long copulation was identical (45.2 min) @aired t-test: t < 0.01, df 12, p = 0.997). Also, the mean time at which short and long copulations occurred did not differ (paired t-test: t = 0.6, df 12, p = 0.56). Neither could we find indications for an alternation of short and long copulations. Summarizing, it appears that short and long copulations occur independently of each other in time and that pairs with short copulations d o not show a lower degree of copulatory activity. When Do Copulations Take Place?
Copulations were slightly more frequent at the beginning of each experiment, but
~
246
A. PITI:RS, A. STRFNG & N. K. MI(:HIEIS
0 2 4 6 8 10 12 14 16 18 20 22
Time of day big. 7: Diurnal frequency dstribution of 137 copulations observed in 47 pairs, summed over 5 d. Long copulations are shaded, short copulations white. The bar o n top indicates day (white) and rught (black)
occurred at a rather constant rate throughout the rest of the observation period (Fig. 6). The wavy pattern (particularly clear in pairs with long copulations only) is a result of a dlurnal mating pattern: of 137 copulations recorded (set A), 60 were observed during the day (16h) whereas 77 were observed during the night (8h) (Fig. 7).This deviates sipficantly from what would be expected if mating activity were random with respect to daylight (x2 = 32, df = 1, p < 0.0001, taking into account that daymight = 2:l). This effect is caused entirely by the long copulations (x2= 48, df = 1, p < 0.0001), not by the short copulations (x2 = 0.05,df = 1, p = 0.83).Copulation duration was not different for matings taking place during the day or during the night (long copulations: Mann--Whitney U = 101 1, n = 96, p = 0.92; short copulations: Mann-Whitney U = 120,n = 41,p = 0.08). The latency between two consecutive long copulations was 37.5 5 3.2 h (range 4100 h; n = 55). Fortynine per cent of all latency intervals were equal to or less than 24 h, with a distinct peak in the 21-24 h range (20%I of all latency intervals). In pairs with both copulation types, the latency after a short copulation tended t o be shorter than that after a long copulation (18.72 3.3 h versus 30.9 ? 6.2h; Mann-Whitney U = 174, p = 0.055). Cocoon Production and Mating Activity
Pairs that had more long copulations also produced more cocoons (Fig. 8). The possibility that only one partner had produced most of the sperm and the other most of the cocoons, seems an unlikely explanation for this finding because a single D. polycbma only rarely produces three or more cocoons per week per individual (unpubl. data) and because the histologcal data (see above) already indicated that mutual sperm transfer is the rule. This relationship did not exist for short copulatiom (R, = 0.325, n = 19, p = 0.17).Pairs with at least one short copulation did not produce fewer
Maung Behaviour in a Hermaphroditic Flatworm
6
I
247
R, = 0.38,n = 55,p < 0.01
0
1
2
3
4
5
6
No. long copulations f,k.X: Relationship between number of cocoons produced over a 5-d period and the number of long copulations in the same time period. When more than one pair had the same x and J values, thc number of pairs is indicated. Pairs with more long copulations produce more cocoons; n = 55 pairs, R, is a Spearman rank correlation coefficient
cocoons than pairs in which short copulations were never observed (Mann-Wkitney U = 229.5, p = 0.41).
Discussion We showed that Dngesia polychroa copulates frequently under laboratory conditions. As in 19.dorotocepha/a (JENKINS & BROW 1964), pronounced courtship behaviour is absent, but we cannot rule out cryptic precopulatory assessment based on chemical cues. This would explain why the animals were often in close proximity for some time before copulating. We found no indications for a distinct role pattern of an ‘active’ and a ‘passive’ animal, as known from D. gonocephulu (C. VREYS,E. SCHO(:KAIIN~;TON, productivity of Lhgesia pobcbtua (Turbellaria: Tricladtda). Ecology 58, 109-1 18. BROWN, H. M., ITO, H. & OGDEN, T. E. 1968: Spectral sensitivity of the planarian ocellus. J. Gen. Physiol. 51, 255-260. BLXR,A. 1928: Zur Fortpflanzungsgeschichte der Siifiwassertricladen. 2001.Jahrb. Aht. f. Syst. 33, 595636. CHVNG, D. J. D. 1987: Courtship and dart shooting behavior of the land snad He/ix+nu. Veliger 30,2439. VAN DI'IVENRODEN, Y . A. & TER MA.AT, A. 1988: Mating behaviour of Ifymnaeustagnuh. Malacologa 28, 5% 64. FISCH~R, E. A. 1080: The relationship between mating system and simultaneous hermaphrdtism in the coral reef fish Ifpplertms nigricans (Serranidae). Anim. Behav. 28, 62-33, _ - 1984 Egg trading in the chalk bass, Jerunus torftgunrm, a simultaneous hermaphrdte. Z. Tietpsychol. 66, 143-151. FISCHISCHURIGER, W. & CIAI.SNIT/.F.R, E. 1984 Bursa, bursa canal and female antrum of Dugesia h@na (Plathelminthes, Tncladida). Zoomorphology 104, 3 8 6 3 9 5 . GIL.STI, F. & ANDREINI, S. 1988 Morphological and ethological aspects of mating in two species of the family Helicidae (Gastropoda Pulmonata): 77xbu pzsana (Muller) and Helix +Mia Born. Monitore 2001.Ital. (N.S.) 22, 331-363. HOFMANN, E. 1923: Uber den Begattungsvorgang von Ariuntu arbustonrm L. Jenaische Zeitschr. Natunviss. 59, 36.3-400. HUMMON, M. R. & HLWMON, W. D. 1992 Gastrotricha. In: Reproductive Biology of Invertebrates. Vol. V. Sexual Differentiation and Behaviour (AlXYoDI, K. G. & Aoruonl, R. G., eds). Wdey & Sons, Chichester. pp. 137-146. HYMAN, L. H. 1925: The reproductive system and other characters of Plonatiu ubmforpbulo (Woodworth). Trans. Am. Microsc. SOC.44,51-89. _ - 1951: The Invertebrates: Platyhelminthes and Rhynchocoela, The Acoelomate Bilateria (Vol. II). McGraw-Hill. New York.
Mating Behaviour in a Hermaphrdtic Flatworm
251
JENKINS, M. M. & BROWN,H. P. 1964: Copulatory activity and behavior in the planarian Dugesia ahton@& (Woodworth) 1897. Trans. Am. Microsc. SOC.83, 3 2 4 0 . JENSEN, K. R. 1986 Observations on copulation in two species of Ebsia from Florida (Opisthobrancha: Ascoglossa). Ophelia 25, 25-32. KEARN,G. C. & WHIT~NGTON, 1. D. 1992: Diversity of reproductive behaviour in platyhehinth parasites: insemination in some benedeniine (capsalid) monogeneans. Parasitology 1 0 4 , 4 8 9 4 9 6 . LASERRE,P. 1975: Clitellata. In: Reproduction of Marine Invertebrates. Vol. 111. Annelids and Echiurans (GIESE,A. C. & PEARSE,J. S., eds). Acad. Press, New York. pp. 215-275. LEONARD, J. L. & LUKOWIAK, K. 1984 Male-female conflict in a simultaneous hermaphrodite resolved by sperm tradng. Am. Nat. 124, 282-286. - _ 1991: Sex and the simultaneous hermaphrodite: testing models of male-female conflict in a sea slug, Nauanax inemzs (Opisthobranchia). Anim. Behav. 41, 255-266. LONGIXY, A. J. & LONGI.W,R. D. 1984: Mating in the gastropod mollusk Aeo(idapopillosa behavior and anatomy. Can. J. Zool. 62, &14. MAIXCHA, J. 1992: Annelida - Hirudmea. In: Reproductive Biology of Invertebrates. Vol. V. Sexual K. G. & ADNODI,R. G., eds). Wiley & Sons, Chichester. pp. Differentiation and Behaviour (ADIYODI, 237-248. REEVE,M. R. & WALTER, M. A. 1972: Observations and experiments on methods of fertilisation in the chaetognath &$a hirpida. Biol. Bull. 143, 207-214. RmE, H. 1995: Mating behaviour of Demceras mdnae GROSSIJ & LCPU,1965, and D. praecox WIKTOR, 1966 (Pulmonata, Agnolimacidae). J. Moll. Stud. 61, 325-330. ROMEIS,B. 1968: Mikroskopische Praparationstechnik. Oldenburg Verl., Munchen. Rui)oi.Pr
