700 PROF. AUTIIUR DEXDY ON 



" showing a sliglit tendency to longitudinal arrangement. "Venti-al 

 " surface pale grey without markings, nearly white. Anterior tip 

 " pinkish brown. Colour of dorsal markings reddish brown, 

 " almost chestnut, in some specimens." In these specimens the 

 tendency of the markings on' the dorsal surface to arrange them- 

 selves in three longitudinal bands on each side of the narrow, 

 mid-dorsal, dark stripe, is more marked than in the Maria 

 Island sjaecimens. These specimens also do not attain so large a 

 size as those from Maria Island, measuring in spirit about 25 by 

 4 mim. 



This species is evidently closely related to the Australian 

 G. qaadrangulata, especially to the mottled Mount Wellington 

 (Victoria) variety of that species. It is, however, a very much 

 more robust form *, being intermediate in this respect between 

 G. quadrangtdata and the Tasmanian G. diemenensis. 



The relationship between these three species requires careful 

 investigation. G. diemenensis, it may be remembered, is charac- 

 terized by the presence of remarkable comb-like copulatory 

 organs, and is accordingly included by von Gratf [1899] in his 

 genus Artioposthia. Such organs have not been observed in 

 G. qiMdrangidata or in G, nichollsi'f, but we must not overlook the 

 possibility that all the specimens of these species hitherto found 

 have been immature and that the peculiar copulatory organs may 

 be developed only in full-grown individuals. The specimens of 

 G. diemenensis in which I observed these organs wei'e collected in 

 February t and March, while those of G. nichoUsi described above 

 were collected in September. The size of the specimens may also 

 depend lai'gely on the time of year. Unfortunately, we do not 

 yet know nearly enough about the life-history of land-planarians 

 to enable us to settle these points. 



Steel has already expressed a doubt as to whether all the 

 specimens originally referred by me to G. diemenensis, and 

 accepted as such by von Graff, really belong to the same species. 

 He himself gives a figure of a specimen of " G. diemenensis," one 

 of a small series of examples from Trevallyn Hills and Table 

 Cape, but I am doubtful whether it leally belongs to that species — • 

 or yet to G. nichoUsi, which it seems to resemble fairly closely in 

 pattern — because the transverse section appears to be of quite 

 different shape, not at all quadrangular, Possibh'', however, the 

 outline given represents the transverse section in life, which is 

 very different from what it is in spirit-specimens. 



in the type-specimens of G, diemenensis from Mount Wellington 

 (Tasmania) there was no narrow median longitudinal stripe on the 

 dorsal surface, and I am inclined to think that the presence of 

 such a stripe in G. nichoUsi may serve as a valid specific distinction. 



* I have, however [1895], described a robust variety of Q. qnadrangulafa from 

 the Blue Mountains, N.S.W. 



t See, however, footnote below. 



J This was one of the Parattah specimens, and as it shows a narrow, mid-dorsal, 

 dark strii^e, I stroiia,'ly sUspect tluit it may be a specimen of G. nicltollsi. 



