110 



the purpose of ascertaining, if possible, how far injuries to plants 

 might go toward producing the obtusilobata forms. These experiments 

 were began as early as May, and continued at intervals of a fortnight, 

 up to the ist of September. During that time some hundreds of 

 plants (the species being everywhere excessively abundant, there was, 

 of course, no danger of exterminating it) in widely different localities 

 were mutilated in every conceivable manner, but no obtusilobata was 

 obtained in any instance. 



These experiments were continued during the next season with 

 the same unfavorable result, and seemed to show that the variation 

 was not due to any interference with the natural growth of the plants 

 by injury, or mutilation of any Uind, whether accidental or ciesigned, 

 and that some other explanation was to be looked for. 



Within a month I have collected three specimens of obtustilobata 

 on plants with well matured, uninjured sterile fronds, and the appear- 

 ance of the plants indicated that they had not been disturbed since 

 they first started to grow in the spring. 



These observations appear to show that the variation is due to 

 some cause within the plant, rather than to any outside influence, 

 however much such influence may assist in developing an impulse 

 already existing. Finding, as I have more than once done, two plants 

 grovying side by side under, precisely the same conditions, one with 

 mutilated and the other Avith perfect, sterile fronds, yet each produc- 

 ing this variation, what other explanation is there than the one found 

 in that disposition to vary which is inherent in the very nature of all 

 plants, and Avhich is likely to manifest itself at any time whenever the 

 right conditions are favorable for such manifestation. 



Take the kindred examples of variation in Osmunda. The laws 

 which govern the variations of Onoclea ought to prevail here as the 

 variation is similar in kind in both genera ; yet, in my own experi- 

 ence, I have never seen a plant with tht frondosa forms that did not 

 have a_ plentiful supply of good healthy sterile fronds, and could 

 never discover any difference in the conditions by which the plants 

 were surrounded in the seasons when they produced the variation 

 and in those when they did not. 



The fact seems to be that there exists in all plants a tendency to- 

 ward variation that will manifest itself occasionally whether the plants 

 are injured or not. That changed conditions will oftentimes assist 

 this pre-existing tendency to develop itself is probable, as we know 

 that plants are so affected, but they cannot be regarded as the real 

 cause^ of these variations which occur quite as often, and, in my 



experience with the variation under consideration, oftener without 

 such assistance. 



Injuries to plants are more likely to produce monstrously aborted 

 forms, and while it cannot be denied that Nature herself in her ef- 

 forts to create varieties, oftentimes produces unnaturally abortive 

 monstrosities, yet her efforts are oftener, and as a rule, directed to- 

 ward the production of improved forms. Now the obtusilobata form 

 ot Unocleaxs not m any sense an abortion, but in its most perfect 

 state IS an interesting and elegant form. Whether it is to be considered 

 as evidence of an effort on the part of the species to create a variety 



