THE PLANT WORLD 105 



there also regarded as lia\'ing good specific characters. The additional 

 fact that many American forms seemed to reoccur in their wild state 

 without change or variation, led several years ago to a careful field 

 study of various forms by several persons in different States. The re- 

 sults have been surprising. In place of the ten species, it is now cer- 

 tain that we have at least 120 species, jnerely on the Atlantic coast. 

 More than eighty species have already been described from this region, 

 while imperfect material at hand indicates that there are probably half 

 as many species yet undescribed. 



While the number of species is undoubtedly so large, the number 

 in any local flora, of an area of about 100 square miles, will in few cases 

 exceed sixteen or seventeen. This is due to the great number of local 

 species, or species with limited distribution. In each of the eastern 

 States there will probably be found species which do not extend beyond 

 its borders. 



Instead of the species being variable and intergrading, as has been 

 so long thought, there is probably no large genus in which the specific 

 characters are more firmly fixed, less variable, and so easily described 

 as in the thorns. What has retarded the study of the genus more than 

 all else, is the habit of collecting flowering and fruiting specimens from 

 different plants, which would often be different species, and distribut- 

 ing under the same label, so that the species in most large herbaria are 

 presented in an almost inextricable confusion. As most of the fruit is 

 collected green, all distinctive shape, color, and the characteristic dis- 

 position of the calyx lobes is lacking. 



In studying the forms of a locality, the first essential is to mark 

 the plants from which specimens are collected, so that flowers, mature 

 foliage and ripe fruit will all be from the same plant. It is necessary 

 to have complete material, flowers, foliage and fruit, since some species 

 are with difficulty separated while in flower, the characters which distin- 

 guish them being most evident in the mature foliage or fruit. In many 

 species the foliage is very similar, and the distinctive characters lie in 

 the flowers or fruit. This adds to the difficulty of identifying speci- 

 mens in immature fruit. 



In the inflorescence it should be noted whether it is cymose or 

 corymbose; simple or compound; the usual number of flowers; the 

 amount of pubescence; whether the flowers are erect or nodding and 

 their size when expanded; the size and number of bractlets, and if they 

 are glandular; whether the calyx is obconic or cup-shaped, and its 

 pubescence. 



The calyx-lobes furnish some excellent characters: their size and 

 shape, whether serrate or entire, and if stalked glands are present; 

 the persistence of the lobes on the mature fruit, and when persistent, if 



