November 6, 1914] 



SCIENCE 



653 



may be arranged in an isomorplious series. 

 In other words, certain genera as regards the 

 hemoglobins are isodimorphotis and others 

 isotrimorphous. 



6. When orders, families, genera or species 

 are well separated the hemoglobins are corre- 

 spondingly markedly differentiated. For in- 

 stance, so different are the hemoglobins of 

 Aves, Marsupialia, Ungulata and Bodentia 

 that there would be no more likelihood of con- 

 founding the hemoglobins than there would 

 be of mistaking the animals themselves. Even 

 where there is much less zoological separation, 

 as in the case of the genera of a given family, 

 but where there is well-marked zoological dis- 

 tinction, the hemoglobins are so different as 

 to permit readily of positive diagnosis. When, 

 however, the relationships are close the hemo- 

 globins are correspondingly close, so that in 

 instances of an alliance such as in Oanis, 

 Vulpes and Urocyon, which genera years ago 

 were included in one genus (and doubtless 

 correctly) the hemoglobins are very much 

 alike, and in these cases they may exhibit 

 closer resemblances than may be found in 

 general in specimens obtained from well-sepa- 

 rated species of a genus. 



So distinctive zoologically are these modified 

 forms of hemoglobins that we had no difficulty 

 in recognizing that the common white rat is 

 the albino of Mus norvegicus (Mus norvegicus 

 albus Hatai) and not of Miis rattus, as almost 

 imiversally stated, and that Ursidse are related 

 to Phocidse (as suggested by Mivart 30 years 

 ago), but not to Canidse, as stated in modern 

 works on zoology. Moreover, we were quick 

 to detect errors in labeling, as, for instance, 

 when a specimen marked as coming from a 

 species of Papio was found to belong to one 

 of the Felidae. Generic forms of hemoglobin 

 when obtained from well-separated genera are, 

 in fact, so different in their molecular struc- 

 tures that when any two are together in solu- 

 tion they do not fuse to form a single kind of 

 hemoglobin or a homogeneous solution, but 

 continue as discrete disunited particles, so 

 that when crystallization occurs each crystal- 

 lizes independently of the other and without 

 modification other than that which is depend- 



ent upon such incidental conditions as are to 

 be taken into account ordinarily during crys- 

 tallization. Thus, the hemoglobin of the dog 

 crystallizes in rhombic prisms which have a 

 diamond-shaped cross-section; that of the 

 guinea-pig in tetrahedra; that of the squirrel 

 in hexagonal plates; and that of the rat in 

 elongated six-sided plates. When any two of 

 these hemoglobins are together in solution and 

 crystallization occurs, each appears in its owBi 

 form. Such phenomena indicate that the 

 structures of the hemoglobin molecules arfr 

 quite different; in fact, more differentiated 

 than the molecules of members of an isomor- 

 phous group of simple carbonates, such as the 

 carbonates of calcium and magnesium whicb 

 when in separate solutions crystallize in 

 rhombohedrons whose corresponding angles 

 differ 2" 15', but which when in molecular 

 union, as in the mineral dolomite, crystallize 

 as a single substance which has an intermedi- 

 ate angle. 



Upon the basis of our data it is not going 

 too far to assume that it has been satisfactorily 

 demonstrated theoretically, inferentially and 

 exjierimentally that at least this one substance 

 (hemoglobin) may exist in an inconceivable 

 number of stereoisomeric forms,^ each form 

 being peculiar to at least genus and species 

 and so decidedly differentiated as to render the 

 " hemoglobin crystal test " more sensitive ia 

 the recognition of animals and animal rela- 

 tionships than the " zooprecipitin test." 



Subsequent to the research referred to in- 

 vestigations have been pursued in the study 

 of hemoglobins from various additional sources, 

 especially from representatives of Primates, 

 with the result in the latter case of finding 

 indubitable evidence of an ancestral alliance 

 of man and the man-like apes. 



More or less elaborate studies by crystallog- 

 raphic and other methods have also been made 

 with other albuminous substances and with 

 starches, glycogens, phytocholesterins, chlo- 



2 Even df we assume that the different forms 

 are not, strictly speaiing, stereoisomers it must be 

 admitted that hemoglobin exists in forms that are 

 specifically modified in relation to genera and 



