CHAPTER III. 

 ON THE CLASSIFICATION OF THE ROTIFERA. 



FOUR attempts have been made to improve upon Ehrenberg's classification : viz. that of 

 Dujardm in 1841, of Leydig in 1854, of Dr. S. Bartsch in 1877, and of Herr K. Eckstein 

 in 1883. I do not intend to discuss here the various merits and faults of these 

 five systems ; it is enough to say that they all seem to have the fault of needlessly 

 bringing together animals that are different in structure, while separating others that 

 closely resemble each other; I say "needlessly" because perfect classification appears 

 to be an impossibility, except at that fleeting stage of our knowledge when none but 

 the commoner genera are known to us. These usually differ from each other in a marked 

 fashion, the very fact of their wide-spread co-existence being perhaps due to their differ- 

 ing so as not to interfere with each other. 



When, however, continued search has brought to light the rarer forms, these usually 

 prove to be links between the more common ones ; and then the troubles of the classifier 

 begin. For these strange forms, which are the delight of the naturalist, are the classifier's 

 despair. Do what he will, no system that he can devise will put into Nature those 

 sharp divisions and well-marked gaps that are so dear to him, but of which she knows 

 nothing. 



Nature has but one law, that of infinite variety ; and the utmost that the classifier 

 can do is to group his animals as well as he can round certain typical forms, content to 

 have the symmetry of his plans and the sharpness of his definitions marred by forms 

 that perversely bear the characteristics of two or three of his types, in nearly equal 

 proportions. 



He may take comfort, nevertheless ; for, even if he had been able to invent a 

 thoroughly satisfactory classification, it is from the nature of the case written in sand- 

 He can never say as he throws down his pen : 



Exegi monumentum sere perennius ; 



for it is almost certain that the fresh discoveries of the next ten years will require his 

 work to be re-cast ; and no higher praise could be given to Ehrenberg's system than 

 that, in spite of new discoveries and its own obvious faults, it has reigned alone for nearly 

 five times the usual period. 



The Eotifera may first be divided into four natural orders, according to their modes 

 of locomotion, and the structure of the foot. The first of these ideas appears in 

 Dujardin's classification, and the second in Leydig's, and they are both excellent; for 

 there are Eotifera that swim by means of their ciliary wreath, and skip by the help of their 

 arthropodous limbs ; Rotifera that swim only with their wreath ; others that swim and 

 creep like a leech ; and lastly, some that, when adult, are stationary. Moreover, in three 

 of these four orders there is only one form of foot in each order, and that form is unlike 

 those of the other two ; and although in one order there are more forms of the foot than 

 one, still they are all unlike the forms of the other three. 



Nor is this all. The natural character of these four orders is further shown by there 

 being other important points of structure, in which the animals comprised in each order 

 at the same time resemble each other and differ from those of the other orders. 



