54'- EINAR LÖNNBERG, (Schwed. Südpolar-Exp. 



A few conclusions may be drawn from these facts recorded above. It is evident 

 that the members of the family Nototheniidœ do not spawn all at the same time of 

 the year and this is not even the case with the members of the genus Xototlicnia 

 itself. Some of the latter seem to spawn in the antarctic spring or, perhaps, already 

 in the later part of the winter (iV. sima, brevipes, macrocephala, and m. mannorata), 

 others in the antarctic autumn or beginning of the winter {N. misops tiudifroiis, 

 and larseni). A third category which in the winter shows genital organs in a mini- 

 mum of development (as for instance N. coriiccps and gibbcrifrons) inost probably 

 spawn in the summer. Treinatoimis hansoni georgiamts seems to belong to the same 

 category as the first group of Notothenias. The same and all the species of Noto- 

 iJicnia, in which the present author has found ovaries in an advanced state of deve- 

 lopment appear to have comparatively small eggs, about i to i'/^ mm., in the ovary. 

 The more specialised members of the family, on the other hand, as Artedidraco 

 mints and Chœniclithys gitnnari have larger eggs, 2'/2 — 4 mm. Thus, the more 

 primitive species have smaller, the more specialised larger eggs. It is only a pro- 

 duct of a consequent development in the same direction when finally the highly 

 specialised deep sea fish Racovitzaia has acquired a "poche incubatrice" as DoLLO 

 recently has described in his interesting memoir repealedl)- quoted. That the eggs 

 when large are demersal, and not pelagic is certain, but whether the small eggs of 

 the more generalised types (Nolotlioiia, Trcuiatomus) are demersal or not, cannot 

 theoretically be fully decided on the base of the relationship as the experience has 

 taught us that of two nearly allied fishes the one may have pelagic (f. i. Clupea 

 sprattiis) and the other demersal eggs (f. i. C. hareiigus). The size of the eggs 

 cannot with certainty help us to decide this question, when the diameter is less than 

 1'/" mm., although it appears probable that eggs which have attained that size al- 

 ready in the ovary may be demersal. There are, however, other circumstances 

 which speak more strongly for the demersal nature of the eggs of Nototlicnia and 

 Trematomus, viz. the variable spawning time compared with the climatic conditions. 

 The variable spawning time seems to indicate that one time of the year should be 

 as favourable as the other for the development of eggs. This might, within certain 

 limits, be true for demersal eggs on the bottom of the sea, but not for pelagic eggs 

 within the true Antarctic region, where the sea, to a great extent, during the winter 

 is covered with thick ice. Therefore, when we have seen that such species as Xoto- 

 tlienia misops tiudifrons, and N. larseni, which also are found within the true Ant- 

 arctic region, are ready to spawn in the beginning of the antarctic winter, it might, 

 without too great a danger of making a mistake, be concluded that these fishes have 

 demersal eggs. A similar judgement may also be passed about Trcinaloi/iiis han- 

 soni (georgianus). 



