194 MASSACHUSETTS HORTICULTURAL SOCIETY. 



more temperate climates their size and number sensibly decrease, 

 until in colder and extreme northern latitudes they become rare or 

 disappear altogether. 



The total number of species in existence cannot be exactly 

 stated, partly on account of the different views held by different 

 authors as to what constitutes a species, but an approximately 

 correct estimate may be given as about thi*ee thousand. 



In the United States they are in the proportion to flowering 

 plants of about one to forty, this being merely an approximate 

 estimate. The actual number of undisputed species at the present 

 time is 160, all of which are represented in the Society's Her- 

 barium. If to this number Ave add the 450 or more Mexican 

 ferns not found, as j^et, within the limits of the United .States, we 

 shall have not far from 620 or more species of ferns belonging to 

 our North American Flora, this being a little more than one-fifth 

 of the whole number in existence. 



In former geological periods the relative proportion of ferns was 

 apparently much greater than at the present time, nearly four 

 hundred so-called species of fossil ferns having been described 

 from the coal regions of the United .States alone. But some of 

 those determinations are, it seems to me, to be received with a 

 great deal of caution. The best pteridologist will not alwaj's 

 venture to name a living plant from the insutHcient data of a 

 mere fragment, or from a single frond, if it be imperfect, know- 

 ing well from observation on living material constantly passing 

 under his eye, how great is the tendency on the part of living 

 species to vary, even in their most important characters ; how 

 much greater then must be the difficult}' of determining species 

 from the too often unsatisfactory data of the geological period, 

 Avhen oftentimes one can have nothing more than a simple tracing 

 of a mere fragment in the rock to judge from. Even the one 

 character, which of all others might be considered as the most 

 reliable, i. e., the venation, or nerve-structure, is known to vary 

 greatl}' in living species ; so that if we were to apply the same 

 method to living species as is applied to fossil specimens, we might 

 easily swell the number of our ferns to much greater proportions 

 than we are now willing to give to them. , 



I propose now to give you a partial review of the life-history of 

 a fern, as distinguished from that of a flowering plant, in order 

 that you may understand more clearly the differences between 



