( 452 ) 



identity and the specific distinctness of the specimens are certainly onr deduction, 

 and the variation of the species thns erected is also onrs. It is not rarely that one 

 meets with diagnoses of species which give the average of some character of the 

 individnals— for example, the average size — which perhaps is not fonnd in any of the 

 specimens measured, or if found may occur very rarely compared with the greater 

 abundance of large and small individuals.* Most deceptive are those diagnoses which 

 contain statements like these : " Colour brown to black ; size 50 to 60 mm. ; 

 habitat India to New Guinea." In such diagnoses the facts are veiled, and we are 

 easily deceived by taking the diagnosis as being the record of facts, while it is 

 a mere statement of an opinion. The erroneous view expressed by Romanes t that 

 geographical races are less abundant among animals than among plants, and 

 Pagenstecher's view J that moths do not vary to any extent according to locality, are 

 the consequence of such deceptive statements on the side of the diagnosticists. The 

 description of a species or variety, therefore, ought to be a pure statement of facts ; 

 as said above, the facts which the diagnosticist deals with are the characters of 

 individuals ; a pnre statement of facts, with the exclusion of any statement arrived 

 at by reasoning, we should have when the characters of the different individnals 

 were recorded in such a way that from the description it would be plainly visible 

 which characters belong to each single individual. When this is done, the state- 

 ment of our opinion as to the specific identity of the specimens, the variation and 

 distribution of the species, etc., cannot aflect the facts, and, therefore, cannot do 

 much harm, even if our opinion should be wrong. Hence we take it that the 

 description of a species or variety being intended to be a statement of facts, not of 

 conclusions, ought to be the description of one individual to the characters of which 

 the different characters of other individuals are so annexed that a mistake as to 

 which individual a respective character belongs cannot occur. That specimen round 

 which the others are grouped in the description is the type-specimen of the 

 description, and as the description is the description of the species or variety (as far 

 as the individuals of the species or variety are known at the time), in the same 

 sense as a figure of an individual is meant to represent the species or variety, that 

 specimen is correctly called type-specimen of the species or variety respectively. 



Besides the pure record of morphological facts, the diagnosticist has to draw 

 inferences from the facts ; and as the recorder of facts ought to know the facts best, 

 the conclusions the diagnosticist arrives at ought to be generally correct if the method 

 of reasoning is correct. The inferences which concern us here are such as to the 

 specific or non-specific distinctness of groups of imlividuals, and hence we shall 

 restrict our discussion to this kind of conclusions. 



If we received a bird of Paradise with conspicuous ornamental feathers, even 

 if the species were quite unknown to ns, we should at once pronounce the specimen 

 to be a male, thougli we know nothing about its having been a physiological male ; 

 and if the quills of the ornamental feathers were surnmndod by a horny sheath, we 

 should conclude that the individual was not yet in full iilumage. As Dr. Martin 

 has succeeded in breeding tailed and tailless females of Papilio memnon from the 

 eggs of one female in Sumatra, we must conclude that also in other districts where 

 the two female forms occur botli can be produced by each of the two. What is 

 found to be true in a number of cases we are bound to conclude to be true in all 



• See Bateson 4: Brindley, P. Z. S. 1892. pp. .'iSn ff. 



t Darwin and Affrr Sanrin, I^ondon, II. 18'Jo. p. 209. 



t Jatirb. Vtr. .VaM. 1896. p. 158. 



