332 MONOGRAPH OF THE GENUS SYNTHEMIS, 



Probabl}' the ovipositor — especially when it is curved, as in 

 >S'. ^e«c;/u'i (Plate vii., fig. 18) — is still of some use in coatrolliug 

 the escape of the exuding egg-masses This may account partly 

 for its I'etention in so many species, long after its original use 

 Las gone. 



In *S. Jlavoterminata, there is a remarkable variation in the 

 length of the ovipositor, compared with the end-segments, in 

 fully developed females. In Plate vii., figs.22«, h, and c, are shewn 

 the ends of three females in my collection. At first sight it 

 looks as if the ovipositor was very variable itself in length and 

 shape; but I think that this appearance is due mostly to change 

 in relative position, and that if the ovipositor of fig. 2 2c could be 

 cut out and measured, it would be very little larger than that of 

 fig. 22a, which projects only half as far. Still, this is evidence of 

 variability of the structure in this species, and it may be that 

 8. flavoterminata is even now in process of losing a structure 

 already somewhat aborted. Another peculiarity in the ovipositor 

 of tlie same species is, that it carries at its tip a small transparent 

 glutinous mass. This is present in all my specimens (twelve), 

 and therefore is not likely to be a remnant of the sticky sub- 

 stance which holds the eggs together. Its use appears to be to 

 supplement the short ovipositor in holding the egg-masses back 

 from too rapid exudation. 



I have figured in Plate vii., the end-segments of the females 

 of eight S[)ecies. Next to *S'. primigenia, the longest ovipositors 

 aie possessed by the four species S. mac7'OStigma(fig.l7), S. leachii 

 (fig. 18), (S'. eustalacta(Gg.l9), and iS'. tas7)ianica{iig.20}. These 

 represent the most primitive form of the organ still existing in 

 Australia. We then have in S. Jlavoterminataifig.'I'l) a consider- 

 ably shorter and more aborted ovipositor. Passing on, we come 

 to S. regina(^gs,.\Qa and h) — a species very closel}'^ allied in other 

 respects to the >S'. eiostalacta-gvo\x^ — with only a small remnant of 

 an ovipositor left; S. c//«Hi7inc<a(figs.21a and b) with a still 

 smaller remnant; S. claviculata (not figured) with a similar tiny 

 remnant; and finally, to the closely allied group of four species, 

 /?. hrevistyla, S. virgula, S. gutJatn, and S. 9ugra(see (ig.23), in 



