ON THE NECESSITY FOR A NEW MONOGRAPH OF ROSES 41 



works quoted. All the sources of information made use of 

 by Watson require to be checked by a monographer per- 

 fectly acquainted with the species and their varieties. 



R. pimpinellifolia and R. arvensis are about the only 

 ones that have almost completely escaped being confounded 

 with others. 



The experience that I have acquired by a long study of 

 the European Roses, and by the examination of rich 

 materials from the British Islands, warrants me in giving 

 here some advice or explanations as to how certain species 

 liable to such confusion should be examined. 



Rosa eanina, L. 



This species is very common. It inhabits the plains by 

 preference ; its abundance diminishes with altitude, and with 

 latitude northwards. On mountains, and in the more or 

 less boreal regions, it is more or less completely replaced by 

 R. glauca (R. coriifolid}. Its variations are extremely 

 numerous, and a considerable number of them have been 

 raised to the rank of species. To settle the synonymy of 

 the English authors it will be necessary to consult especially 

 herbaria, and not to refer to mere descriptions or even to 

 figures. As regards minor variations, even the study of 

 authentic specimens does not always permit us to recognise 

 exactly what the author had in view under a given name, 

 since under the same name he may have distributed forms 

 very different, though of almost similar aspect. The revision 

 that I have made of the greater number of European 

 herbaria, so far as relates to the Roses, has supplied me 

 with frequent proofs of these errors, made even by the 

 authors of species established on varieties or sub-varieties. 



One of the most frequent cases of confusion is between 

 R. eanina (including R. dumetorum, Thuill.) and R. glauca 

 (including R. coriifolia, Fries), at least when in flower, among 

 the glabrous and the pubescent forms alike. In R. eanina 

 the sepals remain reflexed after flowering, or raise themselves 

 very little ; they are habitually caducous at an early period, 1 



1 In some cases of a very rare and quite accidental kind R. eanina shows 

 itself with sepals erect during part of the ripening period. 



