330 MONOGRAPH OF THE GENUS SYNTHEMIS, 



$. With 8 cut oS' slantingly, 9 and 10 bent upwards, and a 

 long ovipositor Palceosynthemis [S. jjrimigenia). 



He adds a note to the effect that, owing to the great rarity of 

 specimens of Synthemis in collections, he cannot classify the 

 other species. 



Of course, such a division, from a knowledge of two species 

 only, could scarcely be expected to be of any value for the recep- 

 tion of other species. Let us take, for instance, the type »S'. 

 eustalacta, with which Forster should surely have become 

 acquainted before making his attempt. It possesses an ovipositor, 

 though neither so long nor of so exaggerated a form as that of 

 S. primigenia. However, it must go into Palceosynthemis. 

 Hence the original type of the genus cannot now be the type of 

 either siibgenus.* Again, S. regina, a species very closely allied 

 to S. enstalacia, possesses no ovipositor, and would therefore go 

 into Eusynthemis with S. hrevistyla, with which it has very little 

 affinity. 



Apart from any other reason, the excessive rarity of the 

 females of these insects is a strong reason for not employing the 

 female sex-organs in generic subdivision. But a study of the 

 ovipositors of the various species will, I think, convince us that 

 they are only of very limited value in this case. 



First of all — and this is suggested partly by Forster's diagram 

 of S. primigenia 9 — it is necessary to remark tliat great varia- 

 bility exists in the appearance of the end-segments. Females of 

 Synthemis, on emergence, ai-e very flabby, and remain so, in their 

 retirement, for long periods, with not only the three end-segments 

 often imperfectly filled out, but with their whole bodies abso- 

 lutely flat from lack of food. Even if one is fortunate enough 

 to capture a well-developed female (generally one which has been 

 ovipositing), one finds considerable difterences in the position of 



* See International Rules of Zoological Nomenclature, Art. 29. "If a 

 genus is divided into two or more restricted genera, its valid name must be 

 retained for one of the restricted genera. If a type was originally estab- 

 lished for said genus, the generic name is retained for restricted genus con- 

 taining said type." 



