1 1 1 



strands sometimes af^ain divide on similar lines, i.e. after 

 growing some length undividedly. Hence the tassels do 

 not cohere, Cockscomb fashion. The extreme regularity 

 of the point at which this multiplication of the terminal 

 cells takes place is very roiiiarkable. A fern frond grows 

 by development of apical cells and in a divided or 

 decomposite frond, that is, a frond bearing not only side 

 divisions or pinnae, but subdivisions of these, or pinnules, 

 or even further divisions to the extent of pinnulets and 

 pinnuletines, to coin a further necessary diminutive, these 

 apical or formative cells multiply laterally accordingly 

 within the coiled up tips which uncoil as this process 

 continues. This is obviously already a very complicated 

 process, and the wonder is therefore the greater when, as in 

 some of the most marked examples of cristation, we find 

 that at a quite definite stage, all over the frond, these 

 apical cells one and all commence to multiply at a different 

 angle to form tassels, thus finishing their work on different 

 lines to those adhered to previously. 



The fact that in some instances, such as in a form of 

 Ptevis sevvuhita known as Applebyanum, we have had heavy 

 bunch crests formed at the tip of a previous season's frond, 

 only one strand of which had remained green and healthy, 

 and that this crest w^as formed by continuous dichotomous 

 forking throughout the season until a mass of threadlike 

 growth formed a large ball, suggests the possibility that 

 this cre-ting in Ferns may be due to an antavistic tendency 

 for the growing points to spread on Mavchantia-\\ke lines or 

 alternatively to an incapacity in the apical cells to close 

 the circuit, as it were, and form a terminal wail or skin, as 

 they should do normally. This case, however, forms an 

 exception to the general rule of cresting, since the tassels 

 are usually complete when the coil containing them unfolds, 

 and all they do subsequently is to expand to full size with- 

 out any subsequent extension of structure. In a number 

 of cases this faculty is developed more or less irregularly, 

 and grades from single fronds, once divided at their tips or 



