D. B. WARD, M. D. 77 



electric pen and are in the hands of those interested in 

 the subject. The catalogue contains references to 8,000 

 species or more. Since then several thousand additions 

 have been made and I would not be surprised if a com- 

 plete list would contain 14,000 or 15,000 names. 



The nomenclature is in a wildly chaotic state, one single 

 form frequently having been named by from two to a 

 dozen different botanists and soaaetimes, as in the case of 

 Ehrenberg, the same man has named his own species over 

 again, having forgotten that he had seen the form before. 

 Slight, very slight, variations in shape and markings have 

 been erected into separate species, the German observers 

 being particular sinners in that respect. The whole 

 thing makes the study of the diatoraaceae a very puzzling 

 and difficult one. If you send a slide to three different 

 experts for identification the chances are that you will 

 get three different names and all of them correct. 



Eliminating all the synonyms and reducing the mere 

 varieties to their proper species would do much toward 

 simplifying this very tangled skein of names, in many 

 instances of bad Latin and worse Greek. 



But with all this done there would still remain an 

 enormous number of species ; more than any one man 

 would be likely to retain in his memory. The literature 

 of the subject is scattered through periodicals, pamph- 

 lets, and monographs and illustrations of special de- 

 posits, so that a complete library on this subject is very 

 expensive and the work of a lifetime to collect. As in 

 all forms of life, animal or vegetable, monstrosities some- 

 times occur — abnormal forms of all kind. 



Double forms, like Siamese twins, distorted shapes and 

 variations in the markings are found. TJius far, how- 

 ever, as in the study of monstrosities generally, little 

 light has been thrown on the causes of their formation 

 nor has much knowledge been added as to the physiology 

 of the normal forms. 



15 



