22 GENERAL PRINCIPLES OF ZOOLOGY 



these extreme cases, it is true, the hydrids die during or at the dose of 

 segmentation, before the embryo is outlined. 



In the case of animals where copulation is necessary the difficulties 

 of experimentation increase, since often there exists an aversion between 

 males and females of different species which prevents any union of the 

 sexes. Yet we know crosses of different species; e.g., between the horse 

 and the ass; our domestic cattle and the zebu; ibex (or wild buck) and 

 she-goat; sheep and goats; dog and jackal; dog and wolf; hare and rabbit 

 (Lcfus daiii'ini); American bison and domestic cattle; etc.; among birds, 

 between different species of finches and of grouse; mallard and the pintail 

 duck; the European and the Chinese goose {Aiiser ferns and .1. cygnoides). 

 Among the insects, especially the Lepidoptera, the cases are many, but 

 the resulting eggs usually produce larv;E of slight vital force. 



C. Fertility of Hybrids and Mongrels. — Since many hybrids, as the 

 mule, have been known for thousands of years, the criterion is, as it were, 

 pushed back one stage; if the infertility in cases of crosses in many species 

 is not immediately noticeable, yet it may be apparent in the products 

 of the cross. While the products of the crossing ot varieties, the 'mongrels,' 

 always have a normal, often an increased, fertility, the products of the 

 crossing of species, the hybrids, should always be sterile. But even this is 

 a rule, not a law. The mule (which only very rarely reproduces) and 

 many other hybrids are indeed sterile, but there are not a few exceptions, 

 although the number of experiments in reference to this point is very 

 small. Hybrids of hares and rabbits have continued fruitful for genera- 

 tions; the same is true of hybrids obtained from the wild Iiuck and the 

 domesticated she-goat; from Anser cygnoides and -i. domesticits; from 

 Salmo salveliniis and 5. fontinalis; Cyprinus carpio and Carassiiis vulgaris; 

 Bomhvx c\nthia and B. arrindia. 



DlfHculties in Classification. — The final result of all this discussion 

 may be summed up as follows: up to the present time, neither by physio- 

 logical nor liy morphological eviiience has there Ijecn found a criterion 

 which can guide the systematist in deciding whether certain series of forms 

 are to be regarded as good species or as varieties of a species. Zoologists 

 are guided rather in practice by a certain tact for classification, which, 

 however, in difficult cases leaves them in the lurch, and thus the opinions 

 of various investigators vary. 



Change of Varieties into Species. — The conditions alwve discussed 

 find their natural explanation in the assumption that sharp distinctions 

 between species and variety do not exist; that species arc varieties wliicli 

 liavc become constant, and varieties are incipient species. The meaning 

 of the above can be made clear by a concrete case. Individuals of a 



