792 THE POPULAR SCIENCE MONTHLY. 



particular case, because the movement in question is undoubtedly 

 useful, and as such, variations in this direction, be they fortuitous 

 or mechanically induced, would be preserved by natural selection. 

 In other words, an ever-recurring mechanical stimulus is presup- 

 posed even on the theory which works entirely with accidental 

 variations, responding more or less fortunately thereto, while, 

 if the stimulus be a direct cause of the favorable variations, its 

 importance as a factor becomes still greater.* 



Our theory of the origin of the peculiar movement in barberry 

 stamens amounts, then, to this : Stimulation by contact at a defi- 

 nite part of the filament for innumerable generations, increase of 

 the protoplasmic contents by the reduction of adjacent parts, and 

 the usefulness of such a movement at every stage of its develop- 

 ment these three factors, although separately incompetent, have 

 yet in combination been the ones chiefly concerned in bringing 

 about through the agency of natural selection such changes in 

 the protoplasm of the sensitive cells as make its fundamental 

 property of contractility prominent to an extraordinary degree. 



Fertilization being accomplished, the single pistil ripens into 

 a berry. In Berberis vulgaris each of the two ovules ordinarily 

 becomes a hard-coated seed flattened on its inner face by pressure 

 (Fig. 20) in much the same way as happens with the two " beans" 

 in a coffee berry. Sometimes (as in the so-called " male berry " 

 coffee) one of the ovules aborts, thus leaving the other to form a 

 seed proportionally richer in reserve food and correspondingly 

 round in form. Occasionally there may be found barberry bushes 

 producing fruit in which both ovules have aborted, f But accord- 

 ding to Buckhout I such individuals " do not constitute a perma- 

 nent variety, for stoneless barberries are only found on old plants, 

 and it has been proved that young suckers taken from them and 

 planted in fresh soil fruit with perfect seeds." Seed production 

 in this case would thus seem to be a question of the plant's vigor 

 at a given period, and so to be comparable with the case of ordi- 



* The belief that stimuli of the sort described directly induce modifications which are 

 inherited has of late years been advocated by Rev. George Henslow (The Origin of Floral 

 Structures). But before this supposition can be accepted in the present case, we surely 

 require an explanation of how it might be possible for changes induced hi the protoplasm 

 of the mature stamens of a given flower to exert a modifying effect on the pollen grains, or 

 the female germ cells, for inheritance must, of course, proceed from them. The pollen 

 grains being separate and distinct, and the female germ cells fully formed and presumably 

 isolated from surrounding protoplasm at the time of the insect's visit, the difficulty suggested 

 would seem to be a very serious one, and, so far as the writer is aware, not even a plausible 

 explanation on this point has been offered. 



f Sturtevant (On Seedless Fruits, Mem. Torr. Bot. Club, vol. ii, p. 3) cites a number of 

 authors who have noticed this phenomenon in barberries. 



% Treasury of Botany, vol. i, p. 136. 



