SCYPHOMEDUSAE 333 



than by all previous expeditions to the South Atlantic and Antarctic ocean together 

 (compare the lists on pp. 334-5). Dr Kemp informs me that in a great number of 

 hauls, whether open or closed, depth gauges were used, and thus better indications 

 of the depth which the net reached were obtained than in previous expeditions. The 

 method of capture must not be overlooked in a discussion of the range of the bathy- 

 pelagic fauna as has so often been done before. 



Although the important reports of the last 30 years, mostly based on collections from 

 the great marine expeditions (those of the ' Challenger ', ' Plankton ', ' Valdivia ', ' Scotia ', 

 'Gauss', 'Belgica', 'Discovery' and 'Southern Cross', 'Albatross', 'Pourquoi Pas?', 

 'Michael Sars' and 'Arcturus') have resulted in a revision of the classification of the 

 Scyphomedusae, many systematic questions have remained unsettled. In this respect 

 the Discovery material proved suitable, for instance in the case of the different stages 

 of development of Periphylla hyacinthiua, hitherto described as different species, or with 

 regard to the various forms of Atolla wyvillei. Here we have to do with some of those 

 puzzling instances (not uncommon among medusae, as already pointed out by Mayer, 

 1910, p. 517, and Bigelow, 1928, p. 517) where the Linnean system of classification 

 proves to be inadequate to express the true relationship of the different closely related 

 forms; for "as the intermediate or transitional stages prove to outnumber greatly the 

 so-called bonae species, the classification breaks down". 



A study of the recent literature on Scyphomedusae clearly shows the tendency to 

 simplify the system as much as possible and to adopt, instead of the many local varieties, 

 races or forms, only a few well-founded species with great variability and large distri- 

 bution. Previously (as in Haeckel) the indication of a new exotic locality sufficed for 

 establishing a new species. Recent authors (Mayer, Bigelow, and the author himself) 

 take the conception of species in a wider sense. In this way the system is greatly 

 simplified, although this procedure, if driven too far, undoubtedly involves the possi- 

 bility of error. Bigelow, especially, in working out the Arcturus material (1928), has 

 gone very far in this respect in revising the genera Periphylla, Limiche, Naiisithoe, 

 Atolla, Pelagia and Aurelia — in my own opinion rather too radically. He appears to 

 have fallen from one extreme into the other. From a similar point of view, but perhaps 

 less drastically, the author of the present report has tried to carry out the study of the 

 Discovery material. 



The material, especially that of the large series of Periphylla and Atolla, is mainly 

 in a very good state of preservation. The brilliant colours, however, have mostly more 

 or less faded or wholly vanished. Some specimens only of Periphylla hyacinthiua, 

 Atolla zvyvillei and most of the individuals of A. chiiiii show fresh unfaded colours. 

 Vanhoeffen (1903), who took part in the cruise of the 'Valdivia', and Broch (1913), 

 who examined the rich freshly preserved material of the 'Michael Sars', were in this 

 respect in a much more favourable position than the author (see, however, p. 393). 



As regards measurements I wish to point out why I have given as few as possible in 

 contrast with the long tables of most previous authors: in the first place because I 

 attribute but little value to measurements in general, and secondly because I have in 



