rmv method. 



Sliiilios on niiiriiio Oslrni'.ods; 9 



from physiology and oecology as from morphology. Wo have not the right to prefer one 

 character to another. W e must t r y t o give a pic t u r e t li a t is as correct 

 and complete a s p o s s i b 1 e of t h e s p e e i e s a s w (> find it in nature. 

 This programme puts a gigantic task before us, but nature is work. 



It is of course not only from the point of view of the identi fication of species that it is ■'^'^""e reasons for th- 

 desirable to pay attention to as many characters as possible. The descriptions of species are 

 not only useful for a barely certain identification. They are also to enable us to decide 

 the mutual relationships of the forms described. The descriptions of the species form the basis 

 on which in most cases the investigation of the natural system of a group almost exclusively 

 must rest. 



If attention is paid to only a few organs this obviously presupposes that the characters 

 that are not taken into consideration are quite constant or, if they are variable, that their 

 variation is accompanied by correlative changes in the organs that are included in the diagnosis 

 or of which reproductions are given. A constancy or correlation of this kind seems of course, 

 even a priori, very improbable. I myself have observed a great many instances in which it 

 does not exist. 



A good illustration of this is shown bv the two species described below belonging to the 

 svib-family Cypridininae, namely Gypridina. (Doloria) levis and C. (D.) pectinata. — I may mention 

 in passing that these two forms played a considerable part in the development of my studies 

 of this group of animals, as it was during the examination of them that I realized the necessity 

 of departing from the old-established superficial methods of investigation and description. — 

 These two species show a striking resemblance with regard to the length and the type of the 

 shell, the endopodite of the second antenna, the seventh limb and the furca, in other words, 

 those organs to which in the group Cypridini formes (cf. below) attention had hitlierto been 

 almost exclusively paid. I too assumed at first that they were quite identical. Only after 

 the number of species investigated by me was increased and I had observed that there was a 

 great difference between the Ostracods of South Georgia and those of the Falkland Islands — 

 Tierra del Fuego did I undertake a detailed re-examination of specimens from both these regions, 

 paying attention not only to the organs mentioned above but to the other organs as well. It 

 was only then that I discovered that this was a case of two very well differentiated species 

 and that profountl differences were present, especially in the maxilla and the fifth linib, 

 in other words two organs to which practically no attention had formerly been given 

 in this group. 



As a proof of how necessary it is to observe carefully in each form tlie conditions of the 



various characters and not to attempt a premature generalization, some examples may also be 



given, taken from forms treated in the present portion of my work. In the sub-genus Gypridina 



the number of furcal claws is quite constant. In some other sub-genera and genera of the 



sub-family Cypridininae this character is constant in each species, but on the other hand 



it is variable for the sub-genus or genus considered as a whole. Finally in a number of species 



in this sub-family the number of the furcal claws differs not only from individual to individual 



but sometimes even on the two furcal lamellae of the same individual. Similar conditions may 



2 



/.oolnff. hldraR, Uppiuln. 8uppl.-Bd. i. 



