February, 1921. 



SCIENTIFIC AGRICULTURE. 



61 



than legumes? In the corn-belt, for example, what 

 value would result if symbiosis were made possible 

 with the roots of corn? Supposing in Westeni Can- 

 ada we could produce nodules on the roots of wheat? 

 This may seem like a rather wild proposition to tackle. 

 As one man put it, "It is like trying to graft a horse's 

 tail on a pig." 



Upon graduating from college in 1914, the author 

 became research assistant to Dr. T. J. Burrill, 

 then professor emeritus in botany at the Uni- 

 versity of Illinois. Doctor Burrill will be re- 

 membered as the discoverer of pear blight, the 

 first known bacterial plant disease, also as the 

 first to introduce the compound microscope to 

 students on this continent. In Bacteriology as well as 

 in botany he was an international figure. We set out 

 to see what could be done by way of extending the act- 

 ivity of the nodule bacteria to non-leguminous plants. 

 At the outset our scheme did not seem so wild. We 

 knew that certain cross-inoculations existed in nature 

 within the legume family, for example, that the bac- 

 teria infecting alfalfa and sweet clover were inter- 

 changeable. The literature contained several refer- 

 ences where men had caused cross-inoculations to 

 occur which did not exist in nature. For instance one 

 man reported the inoculation of soybeans with alfalfa 

 • bacteria. The third point was that this symbiotic re- 

 lation was believed to exist in certain non-legume 

 plants. This will be treated fully later. 



We set out to learn all we could about the bacteria, 

 the root nodules, and the non-legumiuous plants re- 

 ferred to. We secured pure cultures of nodule bac- 

 teria from all the legumes we could find, both wild 

 and cultivated, and tried them out on all the plants for 

 which we could obtain seeds. Thus we enlarged upon 

 the list of known cross-inoculations that occur in na- 

 ture. We tried various means of lireaking the special 

 adaptations which naturalh^ occur. In other words 

 we tried to bring about unnatural crosses such as had 

 been reported, and others as well. Finally we invest- 

 igated the non-legume plants which were said to func- 

 tion as did legumes. 



Cross Inoculations. — The results of our cross inocu- 

 lation studies allowed us to classify the nodule bac- 

 teria into groups, the members within each of which 

 were interchangeable with their co-members. These 

 groups are as follows : 



Group I. 

 Red Clover, TrifnJium prafriisc. 

 ALsike Clover, TrifoJium hijbridiim. 

 Crimson Clover, Trifolium in-carnatum. 

 Berseem or Egyptian Clover. Trifolium aloxandriniium. 

 White Clover, Trifolium repens. 

 Zigzag, or Cow Clover. Trifolium medium. 



Groiip II. 

 White sweet clover, Melilofus alba. 

 Yellow sweet clover, Melilotus officinalis. 

 Alfalfa, Medicago sativa. 

 Bur clover. Medicago hispida. 



Black medick, or yellow trefoil, Medicago hipulina. 

 Fernugreek, Tri</o)irlla foenvm-firarcum. 



Group 111. 

 Cowpea, Vigna sinensi.s. 

 Partridge pea. Cassia chamaecrisia. 

 Peanut, Arochis Jnjpogoea.. 

 Japan clover, Lrspcdcza .striata. 

 Velvet bean, Muruna utilis. 

 Wild indigo, Bai^fisia iinctoria. 

 Tick trefoil. Desmodium canescens. 



Acacia, Acacia armata. 



Dyer's greenwood, Genista tinctoriu. 



Lima bean, Phaseohis lunatus. 



Group IV. 

 Canada field pea, Pisum sativum arvense. 

 Hairy vetch, Vicia villosa. 

 Spring vetcji, Vicia sativa. 

 Broad bean, Vicia faba. 

 Lentil, Lens esculenta. 

 Sweet pea, Lathyrus latifolius. 



Group V. 

 Soybean, Glycine hispida. 



Group VI. 

 Garden bean, Phaseohis vulgaris. 

 Scarlet runner bean, Phascolus mult if lor us. 



riate IV. 

 Root nodules of Ceanothus amerlcanus. These nodules are 

 not produced by the bacteria infecting leguminous plants, and 

 this plant cannot utilize nitrogen of the air. 



Group VII. 

 Lupine, Lupinus perennis. 

 Seradella, Ornithopus .lativus. 



Group Vlll. 

 Hog peanut, Amphii-<irpa monoica. 



Group IX. 

 Lead plant, Amorpha canescens. 

 Group X. 

 Trailing wild bean, StropJwstyles helvola. 



Group XI. 

 Black or common locust, Robima pseudo-a-cacia. 



One of the reasons for seeking information of this 

 kind was the desire to learn what relation existed be- 

 tween the bacteria and their host plants. In other 

 words were plant relationships correlated with mter- 

 changeabilitv of bacteria? If this were true we had 

 in mind making a trial with nodule-bacteria on plants 

 standing in the closest relation botanically to the 

 legume family. 

 We knew in starting that alfalfa and sweet clover 



