295 



«aiiie factors would assert tliemselves in the develoi^meiit of the 

 j)ure-bred pi'ogenv, producing individuals difPering pei^haps as 

 much as those that huA'e just been described. To define these 

 limits of fluctuation it would be necessary to experiment and to 

 observe in the field. Meanwhile the taxonomist will have to be 

 satisfied with describing and disposing the forms in such a manner 

 that they can be recognised and named with a fair degree of cer- 

 tainty. His task will, in the circumstances, be of an essentially 

 practical, and at the same time preliminary, nature. To treat 

 ■all the modifications he can distinguish in his collections as 

 independent units would be as little justified as to neglect them 

 altogether on the assumption that they are tlie expression of an 

 inherent instability. 



All we can assert at present with respect to the Mexican Haw- 

 thorn amounts to this: — There is a tree in cultivation, known as 

 the Mexican Hawthorn, aud designated variously Crataegus stipv- 

 lacca and C. Tneoticana. It is, in spite of a certain instability of 

 some characters, mainly connected with the foliage, a fairly 

 homogeneous and easily recognisable unit. It has its counterpart 

 in the wild state in Mexico. It is associated there with a number 

 of slight .modifications which cannot, from the very limited 

 material at our disposal, be readily and reliably distinguished 

 from it, but at the same time also with a form which is sufficiently 

 •differentiated to be recognised by the combination of its characters 

 .as something different. This form has been described faid figured 

 ■;is MespiJus pnhescens. 



The differences mainlv affect vewtative characters wdiich 

 are known to be individually unstable, and mav be assumed to be 

 subject to the influence of external factors. Notliing is known 

 from tlie standpoint of horeditj^ and hjdmdisation as to the 

 behaviour of the plants under consideration, and consequently as 

 to their genetic relations. Taxonomically this condition would 



seem to find its rational expression in this way. 



f. Ilumholdtiij occasionally verg- 

 ing towards f. stiptdacea. 



Crataeff us p'lfhescens (ixs species) -\t stipalacea, including slightly 



aberrant modifications of 

 f. stipulacea. 



The first attempt to classify the species of Crataegus subsequent 

 to De CandoUe, was K. Koch's.* Among the forms dealt with in 

 his article he distinguished the following species : (1) C. Loddi- 

 gesiaiia (Mespilus stipulacea Desf,, M. Loddigesianus, Spnch)'; 

 (2) C. mexicana; (3) C liypolasia (C. Lambertianaj Hart,, C. 

 anexicana D. Don, Sweet's Brit. FL Crard, hand aliorum); (4), C, 

 puhescens (Mespilus pubescens, //. B. K,); (5) C. stipulacea (C. 

 stipulosa, Steud.y Mespilus stipulosa, H. B. K,); (6) C. quitensis 

 Benth. In the Appendix to his paper (p. 299) his C. liypolasia 

 is reduced to C mexicana, to which he also refers C. suhcrenata, 

 Benth., a name not previously mentioned by him. I shall have to 

 refer to C. stipulosa, C. quitensis, and C. suhcrenafa later on. As 

 Koch had not seen any specimens of the species mentioned except 



* K. Koch in Yerh. Verein. Beford, d. Gartenbau N. Ser. u (1853) 

 pp. 221-312. 



