PART 5 A MONOGRAPH OF THE EXISTING CRIXOIDS 359 



to find individuals of glacialis associated together in considerable abundance, and at 

 various shines of development, and one would therefore not expect to find isolated ex- 

 amples of young individuals unaccompanied by older ones as is mentioned above. 



It is evident throughout his discussion that Carpenter is not entirely satisfied with 

 the status of quadrata. In his account of glacialis he pointed out that a small specimen 

 of this species dredged by the Triton differed from the adults in the same way that 

 quadrata does, and stated that therefore quadrata was to be regarded as a permanently 

 immature form of glacialis. On the next page he uses the expression "if, indeed, the 

 two species are not identical." 



But he closes with the statement that his present impression is that we have to 

 deal with two distinct species, the smaller of which represents a permanently immature 

 form of the larger. 



Answering Levinsen at considerable length again in 1891, he maintained the dis- 

 tinctness of quadrata and glacialis. 



Rankin (1901) admitted quadrata as a species distinct from glacialis, but was doubt- 

 ful regarding its validity. He says that he finds the differences between the two in 

 the material examined by him to be very slight; but there are 5 specimens from Gran- 

 ville Bay which seem to belong to quadrata. Dr. Arnold E. Ortmann, who collected 

 them, told him that in life there is a distinct difference in their appearance, but that 

 this is less evident in the alcoholic material. The color of quadrata is lighter, and the 

 arms have a less feathery appearance, due to the slightly greater length of the brachials 

 and the consequently greater distance between the pinnules. The character of quad- 

 rata to which Carpenter gives specific weight, that is the shorter P 3 as compared with P 2 , 

 he could not find at all well marked, though its segments seemed, as Carpenter says, to 

 be slightly longer than those of the same pinnule in glacialis. 



Michailovskij in 1903 recognized quadrata as distinct from glacialis. 



In 1904 Grieg discussed the status of quadrata, his studies being based upon a very 

 large number of specimens collected by the V0ringen and by the Michael Sars. 



He found that apparently in the young individuals the quadrate brachials are the 

 most usual, while in the older and larger they are triangular; but between these two 

 forms the intcrgradation is complete, while both may be present in the same individual. 

 In specimens with triangular brachials their length is often as great as their width, even 

 in old well-developed individuals in which the brachials are pronouncedly triangular 

 in shape. 



Regarding the relationships of the earlier pinnules, on which much stress was laid 

 by Carpenter, Grieg found that the relative length of P 2 and P 3 in a number of specimens 

 which he measured from Michael Sars station 10, 1900 (see table 11), which are all young, 

 is as 1 : 0.50-0.81, the majority showing the very proportion which Carpenter considered 

 typical for quadrata, that is, 1 : about 0.57; but there is a regular transition to the condi- 

 tion Carpenter considers characteristic for glacialis, 1 : about 0.80. 



In the specimens from Michael Sars station 87, 1901, of which all but one are old 

 individuals, the relation varies between 1:0.58 and 1:1.19. In most of them P 3 isfour- 

 fiftlis of the length of P 2 , while in some it is even longer than that pinnule; but again 

 there are others in all ways typical glacialis in which P 3 is no longer than it is in quadrata. 



Confirming Michailovskij 's observation regarding the great variability in the 

 relative length and in the number of the component segments in the corresponding 

 pinnules on different arms of the same individual, Grieg finds that the relation between 

 P 2 and P 3 on the five arms of a single specimen varied from 1:0. 58 to 1 : 1.00; in another 



