452 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



So far we have dealt with the main classes only, but it was Lucas's 

 merit to establish three excellent divisions for the orders. These 

 were Pale or Blushed Fruits, Striped, and Whole-coloured Fruits. It 

 will be seen at once that these are good and constant characters, and 

 in later authors' hands they have been given an important place. As 

 the system of Diel-Lucas has not been translated into English, the 

 most recent presentation of it by Dr. Engelbrecht is appended (p. 457), 

 with a typical English fruit in each class and order where possible. The 

 true test of a system is of course its capacity to bring together closely- 

 related varieties, and, judged by this criterion, it cannot be said that 

 the Diel-Lucas system is altogether satisfactory. We find, for instance, 

 in the same group ' Worcester Pearmain ' and ' Margil,' two fruits 

 which are not at all likely to be confused ; and the same may be 

 said of ' Stirling Castle ' and ' Peasgood's Nonesuch,' which are also 

 placed in juxtaposition. It is evident that this system is based 

 too much upon form, and the character of the flesh is insufficiently 

 considered. The Diel-Lucas system remains, however, at the present 

 time the only one which is used to any considerable extent. 



Another writer who built upon Diel was Friedrich J. Dochnahl, 

 author of that wonderful multum in parvo "Die sichere Fiihrer in 

 der Obstkunde." This attempt is the most elaborate yet proposed, 

 and is an effort to base the division upon natural species of Mains 

 from which cultivated Apples have been derived. . Dochnahl also 

 took notice for the first time of vegetative characters as well as the 

 fruits, and it will be readily understood that this makes it difficult 

 to put any condensed account on paper. As the classification was 

 founded, as has been said, upon the botanical species, we find the so- 

 called Crabs and the dwarf Paradise Apples included in the scheme, 

 which may be shortly illustrated as below. 



It will be seen from the above table that Dochnahl bases his 

 early divisions upon the distinct species of Pyrus Mains and his two 

 main tribes, being those supposedly descended from P. M. glabra 

 Koch (Weinlings and Markapfels), and from P. M. tomentosa for the 

 Calvilles and following classes numbered 3 to 10. 



Such a classification, excellent as it is in many respects, suffers 

 from the difficulty that it is not always possible to study tree characters 

 with the fruits in hand, and it is very probably due to this reason that 

 Dochnahl's system, in so many ways an improvement upon that 

 of Diel-Lucas, has not supplanted it. 



In the further subdivisions Dochnahl did not adhere to a fixed 

 plan for each. In the Weinlings, for instance, the calyx and pistil 

 are taken into account. In the Calvilles the persistence of the stipules 

 divides a section, as does also the colour of the leaves, and in some cases 

 the presence or absence of fertile or infertile seeds. In the adoption 

 of pale or blushed fruits, striped or coloured fruits, he follows Lucas, 

 as did most of his successors. The system of Dochnahl is extremely 

 suggestive to the student, but for practical use in naming fruits 

 it must be considered too elaborate. It must be added that the 



