DESCRIITIVE KVA TO THE CHARACTERS OB' FRUITS. 95 



Much ii^e has been iikuIc ot" llu' cluir^clors uffbrdcd by the colors 

 of the fruit, seed, etc., the iiiunes of colors used being those given by 

 Kidgway (Nomenclature of Colors for Naturalists, Boston, 188G). 

 In order to facilitate comparison, each name of a color that occurs in 

 the key is followed by a reference to the corresponding plate and 

 figure in the work cited, thus: Maroon (IV. 2). The giant cells, to 

 which the writer's attention was first called by Mr. AV. T. Swingle, 

 occur in a more or less shar])ly defined zone in the softer part of the 

 flesh, close to the skin. It is really the hard, bright-colored, highly 

 refractive masses of cell contents (probably consisting chiefiy ()f tan- 

 nins) that are here described, 'i'licir characters are taken only from 

 thoroughly ripe fruit in wjiich the cell contents have assumed their 

 final shape, size, and color. 



A word is necessary in regard to the form of the key. The varie- 

 ties are divided, in the " Synopsis of the groups," first into two prin- 

 cipal classes, the dry dates and the soft dates. Each group is then 

 subdivided into a number of smaller groups. The designations 

 selected for the two main groups should not be understood as imjjly- 

 ing that all the dry dates are necessarily hard and thin of flesh or that 

 all the soft dates are rich and sirupy. While the extremes of the two 

 types are very distinct, intermediate forms occur, which make it 

 impossible to draw a hard and fast line l)etween them. P^xceptions 

 occur in both categories, and some of the varieties of which the affini- 

 ties are evidently with the first group have thicker and softer flesh 

 than some of the varieties of the second group. In deciding to which 

 of the two classes a given variety belongs, all of its characters nuist 

 be taken into consideration. It should be explained that the classifi- 

 cation, at least beyond the two main groups, is largely artificial, and 

 is desiffued merelv to aid in identifying the varieties. While in many 

 cases it happens that several varieties that are undoubtedly closely 

 related botanically are brought into juxtaposition in the key, this is 

 by no means always the case. 



The key to the varieties is not of the dichotomous form usually 

 employed by systematic biologists, but is modeled upon that which 

 was introduced by Mr. O. F. Cook in his studies of Myriapoda." It 

 has the advantage of eliminating one species (or variety) at every 

 step. It will be noted that the whole key is arranged in pairs of 

 paragraphs. Of each pair, the first paragraph describes a variety, 

 and the second gives the characters which distinguish all of the same 

 group which follow from the one that has just been separated out. 

 Those characters in the first paragraph that are especially to be con- 



alii a paper published in the Annals of the New York Academy of Science, 

 9. p. 8 (189.5). a key of this style was first used by Mr. Cook. See also Proe, 

 U. S. Nat. Mus., 18, p. 82 (1895). 

 92 



