Vol. XVII. No. 421. 



THE AGRICULTURAL -VEWS. 



181 



MATERIAL FOR GENETIC INVESTI- 

 GATION IN THE TROPICS. 



The study of the inheritance of characters in both plants 

 and animals has now reached the stage where a broad view 

 miy be taken of the subject, in so far as it affects the practi- 

 cal agriculturist. When Mendel's law was rediscovered in 

 1900, it was thought that the application of it to plant and 

 animal breeding would lead immediately to results of the 

 highest ecoDoniic importance. But as time went on it was 

 seen that in respect of many points, where improvement was 

 urgently necessary, the application of Mendel's law did not 

 lead to the desired results. 



One of the most destructive diseases of plants is the 

 rust disease of wheat, which causes an annual lo.ss of many 

 millions of dollars. Ten years ago it was announced that 

 Bitfen, of Cambridge, had discovered that resistance to rust is 

 a recessive. It was straightway thought that it would be an 

 easy matter to send out desirable rust-resistant varieties to 

 replace the older kinds. So far as the present writer is 

 aware, no great change has re.sulted in the varieties of wheat 

 grown by English farmers. The evidence presented by 

 Biffen that resistance to rust is a simple recessive is by no 

 means conclusive, and the work of Nilsson-Ehle has shown 

 ihat the whole matter may be much more comple.x. The 

 latter has found that in crosses between two sorts of high 

 resistance there were produced a number of lines, some of 

 which in F.; proved much more susceptible than either 

 .parent. 



It is of the highest importance to settle once and for all 

 the way in which resistance to rust in wheat is inherited, and 

 to determine how far such resistance is compatible with 

 economic characters. 



So far, little progress has been made in the direction of 

 predicting what will happen when plants are cro.ssed. Some 

 few things we can predict. If a new species of the Legumino- 

 sae were discovered in two forms, one with white Howers and 

 another with purple, it would be fairly safe to say that the 

 purple would be dominant over the white, but we cannot go 

 much further. What is needed more than anything else in 

 the science of genetics is an accurate survey of inheritance 

 in a single genus. By an accurate survey is meant the 

 complete elucidation nf the relation of all the possible 

 characters capable of examination, whether physiological or 

 morphological. In the study of a single character it should 

 not be -regarded as sufficient to generalise on the results of 

 100 F,^ plants and a dozen F;, families. To establish the 

 mode of inheritance of even one character, if it be inherited 

 simply, requires that many hundred F. plants should be 

 studied, and as many F.; families as may be considered 

 necessary to establish the pro)iortions in which occur the 

 three types AA, Aa, and aa. No hypotheses based on F.^ 

 results alone can possibly be accepted as more than tentative. 



The dangers of generalising from insufficient data are 

 illustrated from some work of the present writer which has 

 not reachell the publishing stage. A cross was made be- 

 tween two plants possessing different types of pattern on the 

 seed coat, A and B. The F, was always of type A. In the 

 F., the ratio between A and B was not e-xactly 3:1; there was 

 always a slight excess of type A. Out of 58.5 plants, 485 

 ■were A and 100 were B In families uf fifty plants the 

 ratio was often near the 3:1 characteristic of a single genetic 

 difference. A number of F.; families were grown. Some' of 

 the families bred true to A. The B families were i-omposed of B 

 ■only. Others of the A families segregated into A and B. In 

 the heterozygous families thentioof A to B wis again dis- 



torted by an excess of A, and the ratio of A to B was about 

 49 to 1. The proportion of pure to impure dominants, as 

 revealed by their behaviour in F;j, was not 1 to 2. Ten fami- 

 lies bred true, and twelve segregated. From this and from 

 other evidence it was cncluded that patte.-n A was due to 

 two factors, each of which could produce the pattern acting 

 alone, and the effect of the two when combined was the 

 same as the effect of either one. Further, the two factors 

 are coupled according to some system not yet worked out. 

 Had the results of the .second generation been taken, it might 

 have been supposed that one factor was concerned. 



What are the characters necessary to render a plant 

 desirable for genetic investigation? The desirable characters 

 may perhaps be summarized as follows: — 



(1) It should consist of a large number of different 

 forms which will cross readily, produce a large quantity of 

 selfed seed, and be easy of culture. 



(2) It should not be affected by any destructive 

 fungoid disease, or by any insect pest likely to destroy the 

 plants completely. 



(3) It should come to maturity in a reasonably short 

 time, and be capable of culture at any time of year. 



(4) It should be a perennial. 



(5) It should be capable of being multiplied vegetative- 

 ly. It is perhaps impossible to discover an ideal plant. One 

 of the most promising seems to be the Castor plant {Ridnus 

 cofiimuiHs). Another almost as good is the Pigeon pea 

 ( Ca/'anits iiuikus. ) 



In working out the inheritance of characters in the.se two 

 plants it may be suggested that the best method would be to 

 lay out a piece of ground of, say, ■"> acres, as a permanent 

 genetic experiment. 



There could be, say, a row of 200 of each parent, and 

 a similar number of the F,. Next could follow, say, ten 

 families of F^, each consisting of 100 plants. The F.. fami- 

 lies would each consist of, .say, 100 plants. The adoption of 

 this method would render possible a very thorough study of 

 inheritance in the plant chosen. The character could be 

 followed through three generations at a time, and if any 

 further evidence were required on any particular point, a fresh 

 culture could be obtained without much trouble. 



Perhaps the main idea contained in these notes, upon 

 which it is necessary to lay stress, is that of permanency, 

 and the possibility of repeating work over and over again, if 

 necessary. 



When a set of coherent principles applicable to any 

 plant have been established, and when the relations of genetics, 

 cytology, and chemistry have been elucidated, we may be able 

 to tell the practical man how far it is possible to give him 

 what he stands in need of. Of the science of breeding we 

 have as yet barely scratched the surface. 



S.C.H 



DEPARTMENT NEWS. 



Int'oniiatpion has been receivi'd from Sir Francis 

 Watts, KC.M.G., the Imperial Commissioner of Agri- 

 culture for the West Indies, that he had arrived in 

 New York on his return from Bahamas via Florida. 



Mr. H. A. Ballou, .M..Sc., Entomologist on the staff" 

 of the Imperial Department of Agriculture, has left 

 Hong-Kong on his return to the West Indies from 

 Egypt. When he return,*. Mr. Ballon will have gone 

 round the world. 



