May 6, 1910] 



SCIENCE 



687 



plants, and seem to have an educational 

 value in that they attract students to the 

 garden, where they may unconsciously have 

 botanical knowledge thrust upon them. 

 Certain it is that the freshmen who made 

 voluntary investigation of the Cucurbit- 

 ace£E this last fall have come to appreciate 

 the distinguishing characters of some mem- 

 bers of the order— if one can judge by the 

 number of citrons that were found opened 

 by mistake for watermelons. 



Where possible the primitive wild form 

 is grown to show the improvement which 

 has been brought about under cultivation. 

 Thus, seed has been obtained of the wild 

 tobacco {Nicotiana rustica) of Triticum 

 dicoccoides, recently discovered in Pales- 

 tine, and considered the source of emmer 

 wheat; vines of native grapes and of Vitis 

 vinifera show the sources from which our 

 cultivated varieties have been compounded. 



A third division is devoted to pathology. 

 A few of the great groups of parasitic 

 fungi may be represented, such as corn 

 smut, wheat rust, with its alternate form 

 on barberry, black knot of cherry, etc. A 

 variety of bean susceptible to anthracnose 

 will be grown in a plot adjacent to an 

 immune variety, and a striking demonstra- 

 tion of the value of immune races may be 

 expected. Plots can be sown with a mix- 

 ture of grain and weed seeds and the effect 

 of spraying with iron salts upon the com- 

 peting plants be shown. The "calico" or 

 mosaic disease of tobacco is a convenient 

 type to illustrate a disease which is trans- 

 mitted by inoculation, but which is appar- 

 ently not caused by any living organism. 

 Every other plant touched in a row with 

 an infected leaf will contract the disease, 

 and will form a sharp contrast to the unin- 

 oeitlated individuals left as controls. Non- 

 infectious chlorosis of leaves and chlorosis 

 through gi'afting may be better illustrated 

 in shrubby forms. 



A fourth section of the garden contains 

 specimens to illustrate the laws of variation 

 and heredity. Variations are the building 

 stones out of which the plant-breeder forms 

 his new ' ' creations, ' ' and as such should be 

 well classified. Variation in vigor of 

 growth or in qualities of fruit may be due 

 to inherent characters in the germ, which 

 are more or less hereditary and therefore 

 capable of transmission, or on the other 

 hand they may be the response of the plants 

 to recognizable differences in their environ- 

 ment. In the latter group would come the 

 increased growth due to an increase of 

 available food supply. For an illustra- 

 tion there may be grown plots of tobacco 

 in poorly and richly manured soil to show 

 the effect of abundance and lack of food 

 in the substratum. The contrast between 

 the growth of corn sown separately in hills, 

 and the same plant sown thickly in drills, 

 will show the effect of lack of food brought 

 about by competition. In practise tobacco 

 seed is blown, the light which produce small 

 plants being rejected and the heavier re- 

 served for sowing. Plots of tobacco from 

 heavy and light seed, respectively, may be 

 used to show the variation in adult plants 

 due to the differing amount of storage food 

 in the seed. 



Fluctuating variations about a mean may 

 be shown by sowing seeds from a single 

 parent and comparing the offspring in re- 

 spect to a single character. Plants can be 

 grown to show in how far a selection of 

 these fluctuations may be able to change the 

 characters of a given plant. Indian corn 

 furnishes a good example, since in addition 

 to changes in the percentage content of 

 protein, fat and starch in the grain, other 

 clearly defined characters, such as the num- 

 ber, size and position of the ears on the 

 plant, and the number of grains to the ear, 

 have been shown to be markedly influenced 

 by such selection. 



