240 Inheritance in the Groundsel 



by morphological methods alone. Such methods, valuable and even 

 indispensable though they may be, are nevertheless inadequate for the 

 complete solution of such intricate problems as need investigation in 

 these cases. 



Few weeds, moreover, have been as yet studied by experimental 

 methods ; the classical work in genetics has been done on species and 

 varieties of plants long cultivated in gardens — peas, sweet peas, snap- 

 dragons, stocks, etc. 



Such considerations as these induced me, six years ago, to undertake 

 a genetic study of the common groundsel and its segregates. Although 

 it soon became obvious that I should have to register annually two to 

 three thousand plants, and sacrifice the greater part of each long 

 vacation, the results were of sufficient interest to cause me to extend 

 the work year by year. Although the investigation, as a whole, is still 

 very incomplete, it seems desirable that some of the chief results of the 

 last six years' work should now be placed on record. 



Origin of the Experimental Material. Methods. 

 The Radiate Character RR. 



In the year 1891 or 1S92 I observed and examined hundreds of 

 specimens of a groundsel growing in abundance near Cardiff and 

 Penarth, which was chiefly remarkable for its conspicuous capitula, 

 each of which bore from eight to 13 large ray florets. In January, 1894, 

 Dr C. T. Vachell called attention to this plant at a meeting (if the 

 Biological Section of the Cardiff Naturalists' Society and specimens 

 were submitted to W. B. Hemsley, who reported on them as follows; — 

 "I cannot make anything of this but vulgaris b. radiatvs." I had 

 already expressed the .same view, with certain reservations ; based 

 on the fact, that although these plants were clearly radiate gmundsels, 

 they did not agree well with the description and figure of the Channel 

 Islands plant in Sowerby's Botani/. 



The same groundsel occuis in abundance near Swindon, Cork, 

 and Northwich, where, as at Cardiff, it is frequently accompanied by 

 8. squalidus. It has also been sent to me from the Cambridge Botanic 

 Gardens by Mr I. B. Pole-Evans and Prof Yapp. It has spread widely 

 in the counties of Glamorgan and Monmouth in recent years, and 

 specimens have been received recently through Dr Moss from North- 

 wich, Cheshire, and Freshfield, Lancashire. The only other locality 

 recorded for it in Great Britain appeal's to be Bigbury, S. Devon. It 



