246 



no systematic second crossing afterwards would be 

 practicable even if the progeny were fertile, and the only 

 way would be to sow again from Fi, by itself, and trust to 

 the segregation of character in F., showdng the Mendelian 

 proportions. A further drawback is that ferns are not 

 usually fertile in the first year or two, so that experiments 

 would require a much longer period than, for example, 

 annual peas or other plants which yield a fresh generation 

 every year. In this connection it may be w'ell to record 

 here that an undoubted cross between the percvuciate A. f. f. 

 Victoria and the bristly A. f. f. setigevum was raised by 

 Mr. Birkenhead some years ago, the progeny being of true 

 Victorae form, but bristly like setigerum. A sowing 

 from one of these by the writer yielded a number of plants 

 all quite true to the mixed type, a proof, so far as it goes, 

 that in this case no segregation of character occurred on 

 Mendelian lines. 



C. T. D. 



SCOLOPENDRIUM VULGARE VAR. 

 SAGITTATO-GRANDICEPS. 



It is a well-known fact to those who raise Scolopendrium 

 varieties from spores that the progeny is very apt to vary 

 considerably, and, although maintaining the particular 

 characters of the parent, to present them in varying grades. 

 The above-named form, however, is interesting as forming 

 a marked exception, although from its history it would 

 have been expected that considerable variation would have 

 arisen. In the first place the immediate parent was found 

 wild in Cornwall as a form of " lobatum," a common type 

 of quite normal make save that the tips of the fronds are 

 divided into several. In this case, how^ever, the basal 

 lobes were lengthened and somewhat fanned at their tips, 

 so that it was named sagittato-lobatum. As the wild fronds 

 so characterized sent to me bore spores, by way of experi- 



