242 



SCIENCE. 



[N. S. Vol. IX. No. 216. 



other species which do similar work. They 

 are here, as in the former case, perhaps the 

 principal agents in preventing the undue 

 increase of any one species of plant, but as 

 we find here not an effort of man to combat 

 Nature, as it were, by increasing the growth 

 and spread of one species at the expense of 

 the others, but the exact opposite, so, here 

 also, to a degree we find Nature arrayed 

 against man, and insects thus play by no 

 means the same part in the destruction of 

 weeds that they do in the destruction of 

 cultivated crops. Nevertheless, they have 

 an important function in this direction, and 

 it is safe to say that the benefit which the 

 agriculturist derives from their work iu this 

 way is very great. As long ago as the be- 

 ginning of the century it was pointed out by 

 Sparrman that a region in Africa, which 

 had been choked up by shrubs, pei'ennial 

 l^lants and hard, half-withered and unpal- 

 atable grasses, after being made bare by a 

 visitation of destructive grasshoppers, soon 

 appeared in a far more beautiful dress, 

 clothed. with new herbs, superb lilies and 

 fresh annual grasses, affording delicious 

 herbage for the wild cattle and game. 



In a similar waj' Eiley has called atten- 

 tion to the fact that after the great grass- 

 hopper invasions of Colorado and other 

 Western States in the years 1874 to 1876 

 there were wonderful changes in the char- 

 acter of the vegetation, the grasshopper dev- 

 astations being followed by a great preva- 

 lence of plants which in ordinary seasons 

 were scarcely noticed. It is true that some 

 of these plants were dangerous weeds, but 

 others were most valuable as forage for the 

 halfstarved live stock. Moreover, other 

 plants, and especially short or recumbent 

 grasses, took on a new habit and grew lux- 

 uriantly ; one species, for example, Eragrodis 

 poceoides, ordinarily recumbent and scarcely 

 noted, grew iu profusion to a height of three 

 and a-half feet. 



An important, but not generally realized. 



benefit which is derived from the insects 

 may be mentioned under this head, though 

 not strictly belonging here. Kirby showed, 

 75 years ago, that the insects that attacked 

 the roots of grasses, such as wireworms, 

 white grubs, etc., in ordinary seasons only 

 devour so much as is necessary to make 

 room for fresh shoots and the product 

 of new herbage, in this manner maintain- 

 ing a constant succession of young plants 

 and causing an annual though partial reno- 

 vation of our meadows and pastures, " so 

 that, when in moderate numbers, these in- 

 sects do no more harm to the grass than 

 would the shai'p-toothed harrows which it 

 has sometimes beeu obliged to apply to hide- 

 bound pastures, and the beneficial operation 

 of which in loosening the subsoil these in- 

 sect borers closely imitate." 



AS POLLENIZEES OF PLANTS. 



It can no longer be doubted that cross 

 fertilization is one of the very most impor- 

 tant elements in the progressive develop- 

 ment and continued health of the great 

 majority of flowering plants, and, indeed, 

 that it is with some almost a condition of 

 existence. Opposition to this view, at no 

 time especially strong since the publication 

 of Darwin's great work, has become feebler 

 and more feeble until at the present it is 

 not worth considering. 



Comparative experimentation with self- 

 fertilizing and cross-fertilizing plants, re- 

 peated with many species and genera, have 

 shown a superior growth and vitality on 

 the part of those subjected to cross- fertili- 

 zation of such a degree as to leave not a 

 semblance of a doubt ; while in individual 

 cases self-fertilization has been scientifically 

 shown to even result in a deterioration so 

 marked that it has been compared to poi- 

 soning. 



In this condition of affairs it at once be- 

 comes evident that the good offices of in- 

 sects in this dii-ection are of incalculable 



