243 



species which bears crests or terminal tassels, and yet 

 curiously enough so large a number of both British and 

 €xotic species have sported in that way, and fresh and diverse 

 species are so constantly displaying it under culture or as 

 wild finds, that the whole Fern world would appear to be 

 subject to variation in this particular way. It is even a 

 matter of surprise that in the process of specific and generic 

 evolution, in which it is reasonable to assume " sports " or 

 *' mutations " must have played a part, no crested species 

 should have appeared, particularly as cresting does not 

 involve any reduction of spore-producing energy, the very 

 crests or tassels indeed often bearing spores in profusion. 



Variation naturally assumes all grades, and ranges from 



the comparatively slight departures from the common type, 



which may be regarded as mere individualities, such as in 



the Lady Ferns we have alluded to, up to such marked and 



thorough types as form the elitey and represent the 



material upon which the expert selective cultivator works, 



and from which he has obtained through the spore most of 



the finest forms in cultivation. Many thorough-breds yield 



progeny so true to their own type that no progress is 



made, but it by no means follows that every season's 



sowing will yield the same results. When once the normal 



type is departed from, it may be accepted as a general rule 



that variation will occur again. Some years ago we 



made a sowing from our own find Atliyviiim f. f. cristatum 



Kilrushense, the finest wild tasselled form yet discovered, 



but to our surprise the sowing yielded no crested ferns, a 



subsequent second sowing however yielded nothing but 



thoroughly typical forms, including a qvandiceps. We 



therefore concluded that an error had been made in the 



registration of sowing No. i, but this opinion has been 



greatly shaken by the fact that a third sowing, made last 



season with particular care, is showing a mass of Lady 



Ferns, among which so far we are only able to detect two 



or three slightly crested ones. This would seem to indicate 



that different seasons may differently affect the repro- 



