40 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



is often fragmentary, and no real progress was made until it was 

 discovered that certain of the more persistent structures could 

 be clearly correlated with the more delicate ones; so that, for 

 instance, it became possible in the absence of corollas to identify 

 material which without such correlation was not determinable. 

 In my arrangement of these plants I have emphasized what might 

 be termed "vegetative" parts of the inflorescence, but this is 

 because the characters drawn from these are readily demon- 

 strable, while correlated characters drawn from the flowers and 

 fruit are just as striking but more difficult to utilize in the study 

 of herbarium material. The characteristic corolla- forms, too, are 

 complicated, and often when the type is recognizable at a glance, 

 verbal description proves very elusive, and can not be succinct. 



If my efforts shall aid in bringing order out of the present 

 chaos, it makes little difference to me what rank, whether generic 

 or subgeneric, is accorded to the groups I ha\e come to recognize 

 as natural ones. For me the current genus Utricularia comprises 

 more than a dozen perfectly well-defined genera; but my aim is 

 to show that the groups exist in nature, not to quarrel with anyone 

 about what they shall be called. 



A few words may not be out of place concerning those genera of 

 Lentibulariaceae that have hitherto been distinguished from 

 " Utricularia. The most unrelated of these is Pinguicida. The 

 European species of the two genera are so different in appearance 

 that their affinity does not seem to have been recognized until 

 they were brought together in the Linnaean artificial system; then 

 it became evident that the characters possessed by them in com- 

 mon were not purely accidental, but of a fundamental nature. 

 Extension of our knowledge of the species of the world, however, 

 has in so far broken down the distinctions between Pingidctda 

 and the rest of the family, that I know of only two characters that 

 can be depended upon to distinguish this genus from the others. 

 One of these, purely vegetative, is the possession of true roots, 

 not otherwise known in the Lentibulariaceae; in the other genera, 

 however, other organs, sometimes leaves, sometimes stems, mimic 

 roots so successfully that it is not always easy to prove that they 

 are not what they seem. The other definite character of Pingui- 

 cida, while it is found in the inflorescence, does not reside in either 

 the flower or the fruit, but in the fact that the scapes are always 



