334 



SUMMARY AND CONCLUSIONS. 



with the bears, and probably the interrelation of the whole group, has 

 already been pointed out. In this case the comparison of the axial ratios 

 alone shows similarities, but the form of structure of the crystal and the 

 crystal habit are even more important. Thus, the bear crystals all have a 

 habit of twinning on a prism of 60 and the crystals of the sea-lion show the 

 same habit, each forming trillings on such a prism, which do not resemble 

 any other crystals observed. This "bear-type twin," in so far as our 

 investigations have gone, is found only in the genera Ursus, Melursus, and 

 Otaria. Comparing the axial ratios of the bears with that of the one species 

 of sea-lion examined, we find that the correspondence is not complete, but 

 it involves a | and f ratio, as follows : 



TABLE 54. 



The axial ratios with this 2 : 3 relation are not closer than those of 

 some quite unrelated species, and perhaps are only an expression of the 

 general isomorphism of all hemoglobins that has already been stated, but 

 the fact that the crystals of the genera Ursus, Melursus, and Otaria are all 

 monodinic sphenoidal, a very uncommon crystal class in hemoglobins, 

 indicates a close similarity in the forms of structure in the crystals of these 

 genera; and this is perhaps a better test of relationship than even the 

 habit of twinning. This crj^stal class, the monoclinic sphenoidal, was 

 observed also in the hemoglobins of the earless seals, in the harbor seal, 

 Phoca vitulina, and it perhaps may be the class to which the otter crystals 

 belong, although they were recorded as monoclinic domatic. On comparing 

 the crystals of the harbor seal with those of the otter by axial ratios, a fairly 

 close correspondence is seen. The prism angles are identical, 79, and the 

 angle (3 is 72 for the otter and 75 for the harbor seal. The vertical axes, 

 however, stand approximately in a 7 : 4 ratio with each other, as may be 

 seen by a comparison of the axial ratios: 



TABLE 55. 



That is to say, in the prismatic zone the isomorphism is exact, but in 

 the zone including the base and orthodome the relation of the vertical axes 

 is as 7 : 4. 



Zoologists commonly regard the dogs as being closely related to the 

 bears, but in so far as a comparison of their hemoglobin crystals is con- 



