Philip S. Ward.
Ecology and Life History of the Rhytidoponera impressa Group (Hymenoptera: Formicidae). II. Colony Origin, Seasonal Cycles and Reproduction.
Psyche 88(1-2):109-126, 1981.

This article at Hindawi Publishing: https://doi.org/10.1155/1981/39593
CEC's scan of this article: http://psyche.entclub.org/pdf/88/88-109.pdf, 1188K
This landing page: http://psyche.entclub.org/88/88-109.html

The following unprocessed text is extracted automatically from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

ECOLOGY AND LIFE HISTORY OF THE
RHYTIDOPONERA IMPRESSA GROUP
(HYMEN0PTERA:FORMICIDAE)
11. COLONY ORIGIN, SEASONAL CYCLES,
AND REPRODUCTION
By Philip S. ward'
Department of Zoology, University of Sydney, N.S.W. 2006, Australia
This paper is concerned with colony foundation and with seasonal cycles in brood composition and alate production in the Rhytidoponera impressa group, a species complex of ponerine ants restricted to rainforest and other mesic habitats in eastern Australia and New Guinea.
Life cycle information is most complete for confusa and chaly- baea, and most of what follows refers to those species. Relevant data on the other three members of the impressa group (enigmatica, impressa, and purpurea) are given where available. When pertinent to the discussion, some observations on related Rhytidoponera species outside the impressa group are also included. Collection methods are described in Ward (1981). Most of the data are based on field observations and collections. Where ap- propriate, suspected reproductive females were dissected to ascer- tain the condition of the ovaries and spermatheca. Colony origin
In the Rhytidoponera impressa group there are two methods by which colonies can originate:
(i) from lone, colony-founding winged females (queens), in the manner characteristic of many ants; or
present address: Department of Entomology, University of California, Davis, California 95616
Manuscript received b~, the editor April 15. 1981.



================================================================================

110 Psyche [vol. 88
(ii)
as a result of colony fission or budding (hesmosis), in which one or more mated "workers", accompanied by uninsemi- nated nest-mates, leave the parent colony to found a new daughter nest.
As the foregoing remarks imply, there are two kinds of reproduc- tive females: queens and ergatoid (worker-like) gynes, the latter indistinguishable morphologically from unmated workers. This is the first record of reproductive workers in the impressa group (they are common and well-documented in some other Rhytidoponera) where previous reports suggested that the only functional reproduc- tives were winged queens (cf. Brown, 1953, 1954; Haskins & Whelden, 1965). Mated queens and ergatoid gynes never coexist in the same nest, but they often occur in different nests in the same population (in confusa, chalybaea and impressa). This rat her remarkable dimorphism of female reproductives in the impressa group and the resulting differences in colony structure and genetic relatedness will be examined in more detail elsewhere (Ward, in prep.).
There is little information on the frequency of colony fission in worker-reproductive colonies or on the size of newly-budded daughter colonies. Occasionally small isolated clusters of workers and brood are seen in the field, under stones or in rotten log cavities. Table 1 summarizes the composition of four such clusters in the impressa group, and two from other Rhytidoponera species (tas- maniensis and fulgens). Similar observations were made by Haskins & Whelden (1965) on R. metallica. Note that in the two cases where workers were dissected (Table I), only one individual in each cluster was found to be inseminated. In no instance in the impressa group (or in any other Rhytidoponera species) was a single isolated worker, with brood, located in the field, in contrast to the frequent occurrence of single colony-founding queens (see below). The process of colony fission is observationally difficult to distinguish from the movement of a colony from one nest site to another, and the two events may be inter-related. Table 2 summar- izes observations made on colony movement in chalybaea and in three other Rhytidoponera species (outside the impressa group). In only one instance (maniae) was a single colony observed splitting into two nests, but the same event may have been occurring during the other observations, if some workers remained at the original nest site.




================================================================================

19811 Ward- Rhytidoponera impressa. I1 11 1 Table 1.
Composition of small, isolated clusters of workers and brood (incipient worker-reproductive colonies?) in Rhytidoponera confusa, chalybaea, tasmaniensis, and fulgens.
No.
Species Locality Date workers Brood
confusa
,,
,,
chalybaea
tasmaniensis
fulgens
Royal Natl. Park, 5.xi. 1974 5 eggs, larvae N.S.W.
Pearl Beach, N.S.W. 9.iv.1977 25* 9 eggs, 8 larvae Seal Rocks, N.S.W. 14.vi.1977 13 several larvae Whian Whian State 14.v. 1977 9 several larvae Forest, N.S. W.
nr. Wonboyn Lake, 25.x. 1975 6 none seen N.S. W.
Mt. Koghis, 18.ii. 1977 4* several larvae, New Caledonia one worker cocoon
*workers dissected, one inseminated.
In view of the apparent scarcity of very small isolated clusters of workers and brood (of the size documented in Table 1, i.e. 5-25 workers), it seems likely that colony fission in the impressa group often produces daughter colonies larger in size. (Mature worker- reproductive colonies, i.e. those with alates, contain, on average, about 150 workers.) More field observations on budding are needed; the small amount of information accumulated thus far suggests that nocturnal observations might be rewarding. It is also possible that some worker-reproductive colonies develop from former queen-right colonies in which the queen has died. The origin and development of queen-founded colonies in the impressa group has been more extensively documented. Incipient queen-right colonies have been observed repeatedly in the field (Table 3). Mated queens apparently disperse for some distance, undergo dealation, and search for a suitable nest site (under stones, rotting logs, etc.). Having located shelter, the queen excavates a small cavity, lays several eggs, and rears a small brood of workers, the first of these appearing within about 6 months (3-4 months in lab colonies). Unlike the claustral colony foundation typical of higher ants, queens forage outside the nest for food, and feed their larvae partly on insect prey.
The available field information on incipient, queenright colonies suggests that they are usually founded in the spring and early



================================================================================

Table 2.
Field observations on colony movement and worker transport in Rhytidoponera. Time
Species Date Locality (EDT) Weather Observations chalybaea
chalybaea
chalybaea
victoriae
metallica
maniae
l3.i. 1976 NSW: Sydney 6.00 cloudy, after University
28.ii. 1976 NSW: Sydney
University
6.iii. 1977 NS W:
Balgowlah,
nr. Sydney
28.xi. 1976 NSW: Jct. of
Kanagra
Crk. & Coxs
River
1O.i. 1976 NSW: Gordon,
nr. Sydney
l.xi.1975 NSW: 15 km
N Coombah
p.m. raining much
of the day
6.30 sunny period
p.m. after raining
most of the day
6.15- mild, clear
6.45
p.m.
9.45- overcast,
10.30 mild
a.m.
2.30- cloudy,
3.30 following
p.m. rainy period
c.9.00 cloudy
a.m.
several workers transporting others to nest entrance (crack in sidewalk); all came from the same direction several instances of worker transport (and one of larval transport) from top of a stone ledge to a nest entrance on the side
workers carrying eggs, larvae, cocoons and other workers from one nest entrance at the base of a palm tree to another about a third of the way around the circumference, a distance of less than 1 meter colony moving from one stone to another 1 meter dis- tant; workers carrying egg clusters, larvae, cocoons, and other workers
workers transporting other workers (and in one instance a male) from one nest entrance (in soil) to another, 3 meters distant; worker transport still occurring the following day at 10.00 a.m. workers carrying brood and other workers to one of two new nest sites, 0.6 and 2.5 meters, respectively, from old nest site, the directions at right angles to one another; all nests directly in soil




================================================================================

Table 3.
Field data on 43 incipient, queenright colonies (with 5 20 workers). All Rhytidoponera confusa except the following accessions: 2006 (chalybaea), 2620 (chalybaea) and 2580 (impressa). Brood* Probable
Accession Population Dealate year of
no. code no. Date Female(s) Workers Eggs Larvae Cocoons origin



================================================================================

Table 3 (continued).
Brood* Probable
Accession Population Dealate year of
no. code no. Date Female(s) Workers Eggs Larvae Cocoons origin 808
38 lO.v.75
1 ?
- - - 1974
88 1 2 24.v.75
1 7 0 +
+ 1974
970
34 13.vii.75 1 20
0 + 0 1974
1323
38 5.x.75
1 2
+ + 0 1974
1366 8 9.x.75
1 0
+ 0 0 1975
1548
16 7.xii.75
1 0 + 0
0 1975
2020
25 5.xi.76 1 4
+ + + 1975
1971
3 3 18.ix.76 1 0
0 + 0 1976
1988
30 2.x.76
1 ?
- - - 1976
1996
30 3.x.76
2 0 0 0
0 1976
2006
57 4.x.76
1 0
0 0 0 1976
2022
25 5.xi.76
1 0 0 0
0 1976
2433
22 17.iv.77
1 0 0 0
0 1976 or 1977
2444
2 1 30.iv.77
2 0 0 0
0 1976 or 1977
2547
49 16.v.77
1 0 0 +
0 1976 or 1977
2472
39 1 l.v.77 1 12
+ + 0 1976
2580
93 9.vi.77
1 7
0 + 0 1976
2620
88 1 l.vi.77
1 17
0 + 0 1976
2849
30 8.x.77
1 0
+ 0 0 1977
2850
30 9.x.77
1 0
+ 0 0 1977
* brood: +, present; 0, absent; -, no information



================================================================================

198 11 Ward- Rhytidoponera impressa. I I 115 Table 4.
Composition of 322 confusa colonies, with respect to numbers of cocoons and alates, and month of collection. Large standard deviations are due to variation in colony size, and to the fact that not all nests produce alates in a given season. Sample Size # worker cocoons # alate cocoons # alates Month -(#colonies) mean zk S.D. mean + S.D. mean k S.D. Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
summer, and that development proceeds rather slowly. Of the 17 dealate females collected with eggs or no brood at all, 13 came from spring and early summer months (October-December) and only 4 from the fall (April-May) (Table 3). These findings are consistent with the observation that virgin alates usually remain in the nests throughout the winter, and fly in the spring. Nevertheless the occurrence of a few incipient colonies in apparently early stages of development in April and May requires some explanation: it seems likely that either development was hindered in these colonies or that occasional fall mating flights occur.
Colony foundation in the spring and early summer appears to be the pattern followed inpurpurea: Brown (1954) noted many colony- founding dealate females of this species in October and November on the Atherton Tableland, north Queensland. Forty-one of the 43 incipient colonies listed in Table 3 contained only a single queen. The two instances of primary pleometrosis (colony foundation by more than one queen) both involved colonies in a very early stage of development, without brood. One of these pairs (ace. no. 1996) was brought into the lab, and colony development was monitored. The two queens cohabited peacefully from 3 October, 1976 until the end of December, at which time the colony contained 12 eggs, 8 larvae and 8 worker cocoons. The first worker emerged 2 January, 1977; four days later (after a second



================================================================================

Psyche
[Vol. 88
ALATE
COCOONS
WORKER .........
COCOONS 1 4
.................. :.:.:.:.:.:.:.:.
.................................. ::::>;:;:::::::: ........ ....... .......................................................................................... ....... ........ ::::::::::::::::. ............................................................................................................................. .................................................................................................................. > ...................................................... EGGS
l r r
..................................................... ........ .................................................. ........ .................
........
................. ................
........
.................
........
................. ..............:
.......-.........
........
1 1 I 1 1 1 1 I 1 I 1 I J
SEP OCT NOV DEC JAN
FEB MAR
APR
MAY JUN
JUL AUG
(1 3) (47) (42)
(30) (25)
(13) (28)
(30)
(72) (107)
(32) (39)
Figure 1.
Seasonal changes in brood and alate composition in colonies of confusa and chalybaea, as measured by the proportion of colonies with various life stages. Maximum width (shown at either side of figure) indicates that 100% of colonies contain the particular stage. Because there were no obvious differences between species or between years, data covering both species over 3% seasons have been combined. Figures in parentheses refer to the number of colonies sampled in each month. Total sample size: 479 colonies.
worker had eclosed) one queen was found ousted from the nest and almost dead. The colony (with one remaining queen) continued to develop until artificially terminated 15 months later. Subsequent spermathecal dissections and electrophoretic analysis using allo- zyme markers confirmed that both females were inseminated, and that both had contributed worker offspring to the incipient colony. Seasonal cycle in mature colonies
There are consistent seasonal patterns in the occurrence of brood and alates in mature colonies of the irnpressa group. These seasonal patterns are essentially the same for both queenright and worker- reproductive colonies, except that alates in the latter are pre-



================================================================================

Ward- Rhytidoponera impressa. I I
I I I I I I I I I I 1 I
MAR APR
MAY JUN JUL A.UG SEP OCT NOV DEC JAN
(7) (19)
(20) (24) (27)
(6) (1 3) (39) (20) 14)
(4)
Figure 2.
Maximum larval length per nest (measured in millimeters, with larva in natural resting position), in relation to time of year, for 183 nests of confusa. One larva (the largest) was measured in each nest. Figures in parentheses indicate the number of nests sampled. No data available for February. dominantly male only. Figure 1 summarizes the seasonal changes for confusa and chalybaea, in terms of the proportion of colonies containing various stages of brood or alates, for each month. Data on absolute numbers of worker cocoons, alate cocoons, and alates are given in Table 4 for 322 confusa colonies (similar patterns are shown by chalybaea, but numbers average higher). In confusa and chalybaea there appear to be two peaks of egg production-one in the spring (September-October) and another in the late fall (February-April). Larvae are continually present; those overwintering are small to medium-sized and show little growth until the spring, when development proceeds rapidly (Figure 2). Worker cocoons first appear in October-November, and not until



================================================================================

118 Psyche [Vol. 88
approximately two months later do the first alate cocoons appear (December-January). Adult workers emerge from January until June, while alates eclose over a shorter time period (February- April). Most, or all, of the alates overwinter in the nest, and are released in the spring, possibly in several bursts, since some nests have been found with alates as late as November (and one nest with a single male in early December).
Several points of interest emerge from the foregoing: (1)
There is only one crop of adults produced each year, and alate production is restricted to a limited period of the total time that new offspring are produced. At any given time, the standing crop of new cocoons consists on average of no more than about 30% alates (Table 5). These facts may be relevant to a consideration of control of the sex ratio of investment.
(2)
No cocoons are overwintered, and there is a period of 6 months (July-December) when no new individuals are added to the workforce. At first glance, this would seem detrimental to the increased foraging requirements during rapid larval growth in the spring and early summer. However, because of a time lag between worker eclosion and foraging (callow workers remain in the nest) it may in fact produce an effective increase in the foraging force when it is most needed.
(3)
In the absence of data on sex- and caste-specific growth rates, it is difficult to know whether alates arise from the overwin- tering larvae (hence, from eggs laid the previous season) or from eggs laid in the spring. However one piece of evidence suggests the latter: the discrepancy between the appearances of the first worker and first alate cocoons (two summer months) seems to be too great to be explained by assuming that equivalent-sized overwintering larvae require that extra period of time (and quantity of food) to develop into reproductives. Rather, it would seem more likely that the reproductives develop from spring-laid eggs or alternatively from smaller overwintering larvae.
Not all nests of confusa and chalybaea contain alates in a given season, alate production being associated with larger colony sizes (Table 5). Nevertheless, there is considerable overlap in colony size between nests with and without alates, partly due to the fact that worker-reproductive colonies produce alates at a smaller size (and probably younger age) than queenright colonies (Ward, 1978). It seems likely that a variety of genetic, environmental, and develop-



================================================================================

19811 Ward- Rhytidoponera impressa. I1 119 Table 5. Mean colony size (number of workers) for nests with alates and for those without alates at the time of year (February-September) when winged reproductives are normally present.
Mean no. workers (k S.D.)
Species Alates present Alates absent
confusa 203.1k179.9 (n= 132) 83.9k 65.7 (n=41) chalybaea 270.7k206.2 (~68) 146.7k 122.4 (n=41) mental (ergonomic) factors influence the production of alates. The available information on impressa and purpurea indicates a seasonal brood cycle similar to that of confusa and chalybaea. Nests of the two former species collected in the winter in Queensland generally had small larvae (sometimes eggs), alates, and few or no cocoons (sample of 8 impressa colonies, 16 purpurea colonies). A lowland population of purpurea from near Cape Tribulation, north Queensland, was exceptional in overwintering with mature larvae, as well as worker cocoons and adult alates. Nothing is known of the brood cycle in New Guinea populations ofpurpurea which inhabit much less seasonal environments.
Thus, in Australia at least, four species in the impressa group produce one brood of sexuals a year, most or all of which are overwintered in the nest and released in the spring. This occurs despite contrasting climatic regimes at the north-south extremes of range (summer rainy season in the north, and winter rains in the south) (cf. Brown, 1954).
Collections of enigmatica suggest a similar brood cycle (i.e. small, overwintering larvae; cocoons present only in summer), with one important distinction: alates are usually absent from nests in the winter. Of 13 nests collected in the early winter (April 30-July 1) only one contained alates (all males); on the other hand, four out of five nests collected in the summer (January 12-March 7) contained alate pupae (also all males). The differences are significant (p < .02, two-tailed Fisher's exact test), and suggest that alates fly principally in the fall. If this is so, there would appear to be considerable temporal isolation between enigmatica and its two sympatric congeners (chalybaea and confusa).
Mating Flights
Two pieces of indirect evidence suggest that reproductives of confusa and chalybaea normally mate in the spring:



================================================================================

120 Psyche [vo~. 88
(i)
the proportion of nests containing alates is more or less constant throughout the late summer, fall and winter, dropping rapidly in the spring; and
(ii)
there is a flush of colony-founding females in the early to mid-summer. To the extent that impressa and purpurea share the same seasonal cycle, it may be supposed that their nuptial flights also occur in the spring. Alates of confusa and chalybaea were observed actively dispers- ing or swarming on several occasions in rainforest and urban parkland in the Sydney region. All observations but one (out of 15) were made in the spring (September 15-November 1 O), and the only large-scale mating swarms were seen at this time. Most observations involved congregations of males around nest entrances. On six occasions, isolated male or female alates were observed away from the nest, apparently in a dispersing phase. Spring mating flights were observed for 3 consecutive years (1976-78) in the chalybaea population occurring on the University of Sydney campus. Because of the scarcity of information on this important stage of the life cycle, the 1976 mating swarm is described in detail. This flight took place on 4 October 1976, a mild overcast day with brief periods of sunshine and light rain. At the time observations were begun (10:15 a.m. EST) large numbers of chalybaea alates, mostly males, were observed flying in parts of the University campus. Alates were distinctly concentrated into clusters in tree- shaded areas. Three of these concentrations were examined in detail (Sites A, B and C in Figure 3).
Site A. This cluster was centered about a chalybaea nest entrance between two slabs of sandstone which formed part of a stone wall. Between 10:45 and 11:45 a.m. there were several hundred males within 2 meters of the nest entrance. No alate females were seen. Although males spent most of the time on the ground chasing other individuals, the congregation appeared to be formed by males flying into the site. There were large numbers of workers milling around the nest entrance and most behaved aggressively towards the males, but this did not deter the latter from making repeated attempts to mate with workers (and with other males). Three apparently successful male-worker matings were observed; in each instance the pair was already in copulation when discovered, in a position similar to that described by Holldobler & Haskins (1977) for I?. metallica. The worker dragged the male on the ground for about 30



================================================================================




================================================================================

Psyche [Vol. 88
Later in the evening (10:45 p.m.), lesser numbers of alate males were present, but inactive, on the ground at various locations. Over the next 3 weeks small congregations of males were seen around nest entrances, but never in such numbers or frenzied activity as during the large-scale swarm of 4 October.
The Sydney University population of chalybaea consists princi- pally of worker-reproductive colonies, so the preponderance of males among alates is not surprising. The 1976 mating swarms apparently involved insemination of both workers and queens. Only one mated queen was found, however, and it remains unclear if queens mate predominantly in the vicinity of nest entrances or in separate rendevous sites.
Given the limited number of successful matings observed, it is conceivable that the mating swarm had already passed a peak of activity at the time that observations began (10: 15 a.m.). This is also suggested by the absence of workers in a sex pheromone-releasing posture (as described by Holldobler and Haskins (1977) for R. metallica). Such "calling" workers were observed in lab colonies of chalybaea, where the behavior occurred both inside and outside the artificial nest. The posture adopted was similar to that described for metallica (i.e., head and mesosoma lowered, gaster raised and arched, with tergites exposed). In addition, workers repeatedly rubbed the sides of the gaster with their hind tibiae, presumably facilitating release of pygidial (=tergal) gland pheromone. Such rubbing movements have been reported in Amblyopone pallipes queens (Haskins, 1979) but not previously in Rhytidoponera. A mated worker from one of the copulating pairs observed at Site A was isolated in a modified Janet (plaster-of-Paris) nest in the lab and fed on honey and Drosophila. On 11 November the first egg was seen, and by 21 December there were 2 eggs, 1 larva and 1 worker cocoon. Just before the colony was terminated, in March, 1977, this mated worker had produced three worker offspring (the first had appeared on 19 January 1977). This is perhaps the first record among the Formicidae of colony-foundation by a lone worker. However, as mentioned previously, there is no evidence that single workers found colonies in the field and it appears that they are always accompanied by an entourage of uninseminated workers. The inseminated dealate female, also collected on 4 October 1976, was kept under similar lab conditions for five months. The first worker appeared on 7 January 1977. At time of termination



================================================================================

198 11 Ward- Rhytidoponera irnpressa. I1 123 (March, 1977) the colony consisted of 1 queen, 1 worker, 1 worker cocoon, 1 larva and several eggs.
The following spring, in the morning and early afternoon of 1 October 1977 another large mating swarm of chalybaea occurred on the University of Sydney campus. As before, this consisted mostly of male alates, concentrated into more or less discrete clusters around several nest entrances. Large clusters were situated at Sites A and C (Figure 3), at exactly the same places observed in 1976. No matings were directly observed, but a timid worker which was being mobbed by males was later found to be inseminated. Workers were generally very aggressive towards males, but the latter persisted in attempts to mate. Once again, samples of alates were collected from various sites, of which 97.0% (195) were males and 3.0% (6) were females. These figures are not significantly different from those of 1976.
On October 12, 1978 small swarms (20-30 individuals) of Rhytidoponera chalybaea males were observed at Site C and at several other locations on campus (but not Site A). At 10:15 a.m. males were mostly at nest entrances, apparently in the process of emerging. One alate female was observed; this individual emerged from a nest entrance, and flew off into open sky, ascending rapidly. Similar behavior was observed in males. By 11:30 a.m. many males appeared to be flying into the area, congregations had formed outside nest entrances, and males made repeated attempts to mate with workers.
On the afternoon of the same day two chalybaea queens (one alate, one partially dealate) were seen floundering on the sidewalk in a heavily built-up section of downtown Sydney. Both were unin- seminated. This suggests that alate females may disperse a con- siderable distance before mating.
Colonly Structure and Life Cycle
In most populations of confusa, chalybaea, and impressa, queen- right and worker-reproductive colonies coexist, in intermediate proportions. Despite the likely disparity between mating sites of winged queens and workers, genetic data from electrophoretic studies (Ward, 1978, 1980) reveal no indication of extensive inbreeding or assortative mating with respect to colony type. This is consistent with the observation that brood development and alate production proceed at similar rates in the two colony types, and that



================================================================================

124 Psyche [Vol. 88
release of alates in worker-reproductive colonies occurs syncronous- ly with (or at least in the same season as) queenright colonies. As for the remaining species in the impressa group, only queenright nests are known in purpurea and this species shows a brood development pattern similar to the three others. By contrast, distinct winged queens are unknown in enigmatica (all recorded colonies worker-reproductive), and this species diverges from its closely related congeners by releasing most alates in the fall, although males were found overwintering in one nest. The limited information indicates a possible relaxation of synchrony in the release of ergatoid-seeking male alates, a pattern which would be predicted with the loss of the winged queen caste, especially if the sexual calling behaviour of ergatoid gynes is temporally dispersed. This trend is continued in some other Rhytidoponera species outside the impressa group, in which functional queens are rare or absent, and flights of alates (males) are reported to be highly non- specific with respect to season (Brown, 1958; Haskins & Whelden, 1965; Haskins, 1979). However, since most of the data come from lab colonies of one species (metallica) additional field observations are desirable.
Scattered collections of colonies from different times of the year may give a misleading impression of patterns of alate production. In at least two species of the impressa group, alates can be found in some nests from February to November. Although this superficially suggests aseasonal production of alates, a detailed examination of brood development demonstrated that only one crop of alates is produced each year and that alates are released over a limited time period. Additional field studies are necessary to determine whether brood development in Rhytidoponera species without queenright colonies is less constrained by the need for synchronous alate release. For comparison with the impressa group, such studies would be most appropriately directed towards other species of east Australian mesic forests, in order to minimize climatic and other environmental differences.
In the Rhytidoponera impressa group there are two kinds of colonies, which are distinguished by the type of reproductive female present: queenright colonies with a single dealate queen, and



================================================================================

19811 Ward- Rhytidoponera impressa. I1 125 worker-reproductive colonies in which one or more mated "work- ers" occur in lieu of a queen. It appears that worker-reproductive colonies normally reproduce by colony fission or budding, although information on this process is fragmentary. Queenright colonies are founded by lone queens. Colony-founding queens are most fre- quently encountered in the spring and early summer; such queens leave the brood chamber to forage for food. In mature colonies of confusa and chalybaea, the development of brood and production of alates is highly seasonal (and essentially similar for both queenright and worker-reproductive colonies). One crop of workers and alates is produced each year, the former eclosing from cocoons between January and June, the latter between February and April. Most or all alates overwinter in the nest (along with small to medium-sized larvae), and are released in the spring (September-November). Similar seasonal patterns are shown by impressa, purpurea (in Australia), and enigmatica, except that colonies of enigmatica generally do not retain alates over the winter.
In the population of chalybaea on the University of Sydney campus, mating flights took place in early October for 3 consecutive years. During these flights, flying males became concentrated into clusters around nest entrances where they attempted to mate with workers, with males, and with the occasional alate female. Several worker-male but no queen-male matings were observed in these nest-associated swarms. Like males, queens appear to disperse some distance before mating, and possibly utilize mating sites other than nest entrances.
This work was supported by an Australian Commonwealth Scholarship. Additional support from L. C. Birch and the Univers- ity of Sydney is gratefully acknowledged. I thank D. Feener and A. Forsyth for comments on the manuscript.
BROWN, W. L.
1953.
Characters and synonymies among the genera of ants. Part I. Breviora, 11. 1-13.




================================================================================

Psyche [vol. 88
1954.
Systematic and other notes on some of the smaller species of the ant genus Rhytidoponera Mayr. Breviora, 33, 1- 1 1. 1958. Contributions toward a reclassification of the Formicidae. 11. Tribe Ectatommini (Hymenoptera). Bull. Mus. Comp. Zool. Harvard, 118, 175-362.
HASKINS, C. P.
1979.
Sexual calling behavior in highly primitive ants. Psyche, 85, 407-415. HASKINS, C. P. AND W. M. WHELDEN
1965.
"Queenlessness", worker sibship, and colony versus population structure in the formicid genus Rhytidoponera. Psyche, 72, 87-1 12. HOLLDOBLER, B. AND C. P. HASKINS
1977. Sexual calling behavior in primitive ants. Science, 195, 793-794. WARD, P. S.
1978.
Genetic variation, colony structure, and social behaviour in the Rhyti- doponera impressa group, a species complex of ponerine ants. Ph.D. Thesis, University of Sydney.
1980.
Genetic variation and population differentiation in the Rhylidoponera impressa group, a species complex of ponerine ants (Hymenoptera: Formicidae). Evolution, 34, 1060- 1076.
1981.
Ecology and life history of the Rhytidoponera impressa group (Hymen- 0ptera:Formicidae). I. Habitats, nest sites, and foraging behavior. Psyche, 88: 89-108.




================================================================================