,2 LYCODIN^. 



considers that his specimen (which I have examined in detail p. 28 — 30) fills up the gap between the 

 two forms in this regard. Lonnberg might even have concluded this from the large series of 

 measurements of L. esmarkn\ which Collett published in 1884. Putting these into percentages, they 

 show that the length of the head in L. esniarkii varies from 19,2 — 24 " o of the total length; as the 

 relation in /.. vaJilii according to numerous measurements b\- myself, is 18,8 — 23,8 ",r„ no specific difference 

 consequenth- can be founded on this. Nor does the second important proportion give an\- basis for a 

 distinction; the distance between the snout and the anus for example is in L. esmarkn 38— 42,9° o, 

 in L. vahlii 36,5 — 42,3 "/'o')- 



When Collett further asserts that the .shortness of the row of teeth on the palatal bones 

 distinguishes L. esmarkii from L. vahlii ^ where this row is as a rule longer than that on the iuter- 

 maxillar\-, seldom if ever shorter, he has allowed himself to be misled by I^iitken's erroneous observa- 

 tions; as we ha\e seen (p. 18 & p. 20) the palatal row of teeth in the adult males of L. vahlii is 

 alwa\-s shorter than that on the intermaxillar\-. This character on whose uncertainty Prof. S mitt 

 has already remarked, must therefore also fall to the ground. 



Lasth", Lonnberg remarks that little reliance can be placed on the character, that I., vahlii 

 has only one, L. esmarkii two lateral lines, since Collett has seen traces of a mediolateral lateral line 

 in one of the type-specimens of L. vahlii. and con\-ersely the mediolateral line in L. esmarkii is often 

 defaced. This must however be corrected, as L. vahlii never occurs with a mediolateral lateral 

 line; this specimen, on which Collett has based his statement and which is in realit>- one of 

 Reinhardt's type-specimens, is in no wa>- L. vahlii but belongs to the following species which is 

 pro\'ided with two lateral lines (see more in detail p. 36). 



If now, one wishes to settle the independence of L. esmarkii — just as it has been done above, 

 by consideration of the identical and exclusion of the unrelated elements — as against L. vahlii, one 

 must first and foremost lay stress on the following characters: (i) want of pyloric appendages; (2) the 

 larger nmuber of ra\s in the pectorals; (3) the characteristic colouration; (4) the double (ventral and 

 mediolateral) lateral line. 



They differ from one another also in biological relations; L. esmarkii lives on the whole at 



greater depths than L. vahlii, and feeds chiefly on echinoderms whilst L. vahlii feeds on Crustacea 



and Mollusca. 



Distribution. 



According to Collett, L. esmarkii must be considered a stationary and scarcely a rare 

 fish on the coasts of Finmark; almost all the specimens examined hitherto have been caught in the 

 Varanger Fjord on lines, and at the depth of 150—200 fathoms. Between Norwa>- and Bear 

 Island (73" 3' N.L. 18° 30' E.L.), where the depth was 410 m. and bottom-temperature + 2° C. the 

 Nathorst Expedition caught the \oung specimen (192 mm.) referred to in detail above (p. 28) on the 

 4th of September 1898. 



It was taken by the Norwegian Fisheries steamer Michael Sars in the sunnner of 1902 at 



the following places: Slope between Norwa> and Shetland (62^ 30' N.L. 1° 56' E.L.), depth 



') The apparently smaller variation in L.es>na>'kii arises from the fact that the numbers are based on measurements 

 of only 5 specimens and of these but one only was a young individual. (Appendix: in one of the specimens obtained later 

 ["Michael Sars » 1902] the distance between the snout and the anus amounts to only 37,5 °;o of the total length). 



