130 THE JOURNAL OF BOTANY 



the Rubi cross very freely ; and, if so, of course not only would the 

 hybrids be very various, but each kind of hybrid would vary a good 

 deal, thus adding confusion to the confounded. It has been told 

 me that it is pure theory to suppose any Rubiis to be a hybrid. Is 

 it not just as much so to suppose it to be an independent species 

 when it closely resembles two others ? R. Idans and casiiis are 

 admitted to cross. Why not then all the intermediate ones ? I 

 am not unlikely to be mistaken ; but I fancy I have noted the 

 following things to indicate crossing in the Rubi:— A striking 

 resemblance to some salient feature in each of two other neigh- 

 bouring Rubi combined with an intermediate character on the 

 whole ; a very sparse distribution — an isolated individual occurring 

 here and there, and not large numbers, about the same spot, as 

 commonly happens with the distinct species ; fasciculation of the 

 branches of the panicle ; and sometimes some abnormality in 

 growth. Whether what I take to be probably hybrids are less 

 fertile than other Rubi, I cannot say. I have not observed that 

 they are. In California the fruit-growers have found that certain 

 horticultural varieties of fruit trees are incapable of ^//-fertilization, 

 and that their fertilization can be secured by planting other varieties 

 near them. Is it not likely therefore that the same end may be 

 attained in the same way in nature ? What an endless and con- 

 tinually changing variety this would lead to I 



My close observation leads me to gather that such a variety 

 does exist among the British Rubi, and this suspicion is much 

 strengthened by the partial confusion which all others, as well as 

 myself, appear to get into when we try to relegate every Rubus to 

 some distinct and exact spot in any regular scheme of arrangement; 

 and also by the remarks that have been made in the course of 

 doing so. Is not the provision of a clump of Egregii — very unlike 

 to one another — a clever device to lessen the difficulty ? Never- 

 theless careful attempts to form a complete classification must be a 

 most valuable, perhaps essential, step towards an accurate know- 

 ledge of any difficult genus. There are certain species (or forms) 

 which stand out conspicuously from all the others, and are easy to 

 name and remember. But there are still more numerous inter- 

 mediate forms which resemble, it may be equally, two or more of 

 the leading ones, and still more closely resemble one another with- 

 out being identical. These are the ones that do not lend them- 

 selves readily to classification. The difficulty is increased, more- 

 over, by their being usually far less abundant, and consequently 

 much less frequently seen. Thus the eye cannot get so familiarized 

 with them in the fresh state. My own observation has led me to 

 suspect that this thinness of distribution is the main cause why so 

 many have been recorded for only a very few counties, rather than 

 a want of wide distribution. I have found so many of those 

 supposed to be rare in a very small space near me. It needs 

 tinuous as well as long-continued viewing of the Rubi one is already 

 acquainted with to keep one's eye familiar with their distinctive 

 features. If one studies the comital numbers for the distribution 

 of each species at the end of Mr. Rogers's Handbook, it will be seed 



con- 



