MECHANISM OF PROTEIN HYDROLYSIS AND IMMUNIZATION 49 



intermediate between dog and cat tribes, but, contrary to what 

 might be expected, it is much more closely related to the cat 

 than to the dog. The seal and sea lion, on the other hand, are 

 closer to the dog family than to the cats. Moreover, the seals 

 are somewhat more closely related to the weazel tribe than to 

 the felines. 



The porpoise, which might be supposed to be allied to the 

 seal, is found instead to show close affinities with the ox tribe, 

 and in particular with the pigs. Indeed the porpoise may be 

 regarded as a pig that has taken to the water and perforce be- 

 come carnivorous in diet. It is necessary also to record the 

 rather unflattering observation that the blood of the porpoise 

 shows more pronounced affinities with human blood than with 

 that of most other animals. 



The family groupings among reptiles show close blood rela- 

 tionship between lizards and serpents, and a slightly less close 

 relationship between turtles and crocodiles. The reptiles are 

 more closely related to birds than to mammals. The relation- 

 ship appears to be particularly close between birds and turtles; 

 less close between birds and crocodiles ; the avian relationship 

 with lizards and serpents being still more remote. 



These tests singularly confirm the conclusions of the zoologist, 

 based on study of the anatomical structures of the different tribes 

 of animals ; but the testimony is absolutely independent, the tests 

 being made, as already pointed out, by means of blood alone. 



Indeed the maker of the test may never have seen a specimen 

 of the species whose rank in the organic scale he is determining. 

 The specimens of blood that Professor Nuttall used in his classi- 

 cal series of experiments were collected from a multitude of 

 sources; no fewer than seventy different persons sending speci- 

 mens from different parts of the globe. 



Many of the collectors were hunters, who merely dipped a 

 piece of filter paper in the blood of a quarry and transmitted 

 it to the Cambridge Laboratory. There the discolored piece of 

 paper was soaked in water to produce a clear solution of blood 

 serum. A portion of this solution was placed in a test-tube, and 

 this test-tube put in a rack along with scores of other specimens, 

 each bearing only a number. 



Into each test-tube a small drop of a certain liquid was placed. 

 If the solution in the test-tube became cloudy, the experimenter 

 was able to pronounce definitely that the blood was that of an 

 animal of a certain tribe. It might, for example, be the blood 

 of a tiger or a leopard or a panther or a cheetah; but it could 

 not be the blood of a hyena or a wolf or a dog. 



Again the test might be applied to a blood stain on a hand- 

 kerchief or knife, or on a fragment of wood from a floor or 



