356 



HORTICU LTURE 



September 14. 1907 



AN IMPORTANT NEW BOOK ON 

 PLANT BREEDING. 



Mucb as the practical mau is prone 

 to lacli in respect for the "theoretical" 

 and the "scientific," it is rapidly com- 

 ing to be understood that the highest 

 success in agriculture and horticulture 

 can be achieved in these days only by 

 means of a very thorough knowledge 

 of the nature of plants and animals. 

 After all, science is but the systemiza- 

 tion of the kind of knowledge which 

 the "practical" man accumulates from 

 experience and from tradition. Botany 

 has learned much from the "unscien- 

 tific" plantsman; only it has gone 

 much fui'ther in most directions than 

 he has, and has combined sorts of in- 

 formation which to the grower are 

 either unknown or unrelated. The 

 trouble with botanical science — from 

 the grower's point of view — has been 

 that the theoretical man has cared 

 more for theories than for results. 

 Nowadays this condition is rapidly 

 changing. The two kinds of workers 

 are joining hands, and the good effects 

 begin to be seen. The practical horti- 

 culturist can no longer afford to stand 

 apart. The rule of thumb is out of 

 date. It's too short to measure the 

 distances covered by present-day pro- 

 gress. 



Possibly the most important ad- 

 vances in agricultural and horticultural 

 methods, which mark our era, are 

 those in improvement of plant breeds. 

 The purpose of this article is to call 

 the attention of plantsmen to a book 

 Just published by Hugo de Vries, with 

 the title "Plant Breeding." De Vries, 

 as many readers know, is not primarily 

 a grower or improver of races of 

 plants. But as Professor of Botany in 

 the University of Amsterdam he has 

 been a devoted student of heredity for 

 many years. He rediscovered Mendel's 

 work and showed its importance. He 

 has written the most talked-of book 

 on Evolution which has been published 

 since Darwin. His theory of the 

 change of characters in plants is called 

 the Mutation Theory, according to 

 which new characters are not gradually 

 built up by slow change and selection, 

 but appear suddenly by leaps, or 

 "mutations;" as the practical grower 

 would say, they appear as sports. New 

 races of plants come suddenly, accord- 

 ing to De Vries. Moreover, many, 

 if not most species, according to 

 him, consist of a mixture of 

 races, or little species. In working 

 out his ideas de Vries, like Darwin, 

 has constantly referred to the history 

 of horticultural plants, and moreover 

 has for many years actually experi- 

 mented on heredity in the botanic 

 garden at Amsterdam. Horticulture 

 and agriculture have served him in 

 the cause of science. His latest book, 

 named above, is written in the serv- 

 ice of these arts. It makes clear a 

 most important principle in plant 

 breeding, by the best of all methods, 

 namely, that of concrete example. 



De Vrits describes the work of sev- 

 eral eminent breeders, especially Nils- 

 son and Buib'.;nk, the achievements of 

 the former being probably more im- 

 portant to the race at large, though 

 less spectacular, than those of the Cali- 

 torniaii. The work of the Swedish 

 breeder also has this cardinal virtue, 

 that it has been conducted with 

 system, every step being known and 



recorded, and the principles discovered 

 therefore stand forth as clear as day, 

 so that the work may be repeated and 

 applied at large by anyone. The ma- 

 terial that Nilsson has worked on 

 chiefly — the cereal grains — has been 

 also most difficult of treatment, and 

 the problem, vastly impoitant it is to 

 mankind, could only have been solved 

 in its entirety by the scieniific methods 

 in use at Svalof. 



Nilsson has been Director of this 

 Swedish Agricultural Experiment Sta- 

 tion since ISOO. In ISSti the Swedish 

 grains were found to be deteriorating 

 and a society was formed in the village 

 ot Svalof in southern Sweden, for the 

 introduction, testing, and improvement 

 of the best foreign sorts. The first 

 work of the society was conducted 

 along the lines of the German grain- 

 breeders. In brief the German method 

 was — and is — to select from a field oi 

 growing wheat, oats, or rye, numbers 

 of the best stands showing the largest 

 ears with the most kernels, or mani- 

 festing in a high degree whatevei- 

 quality is sought for. This handful 

 of plants, the elite, becomes the basis 

 of the seed ciop of the next year. Some 

 improvement is expected each season; 

 and in fact in this sort of selection 

 considerable success has been attained 

 in the evolution of improved grains. 

 The seedsmen have always recom- 

 mended the purchase of seed from 

 their own establishments on the 

 ground that only by the continued 

 selection of elite strains could purity 

 and standards of productiveness be 

 maintained. 



This method was practiced at Svalof 

 when Dr. HJalmar Nilsson, the pres- 

 ent Director, took charge, in the year 

 lyOO. His first discovery was the in- 

 adequacy and the uncertainty of the 

 method. Selection resulted only rarely 

 in any marked success. It was found 

 in certain cases that qualities of a 

 particularly desired kind — for example 

 stiffness of stalk in oats — could not be 

 induced by the most persistent efforts. 

 Moreover, Nilsson found that the plots 

 of grain sown from elite strains were 

 not uniform, or pure. The impurity of 

 the strains was, it is now thought, the 

 source of deterioration in crops and of 

 defeat in the experiments. 



Nilsson's first important observa- 

 tion, was that among the thousands 

 of cultures, some few were found that 

 appeared to consist of only one type. 

 As elaborate records of all the original 

 samples of seed has been kept, Nilsson 

 was enabled to ascertain the cause of 

 the purity in these exceptional cases. 

 According to the accepted method each 

 sample had consisted of a certain 

 number of ears which were as similar 

 to one another ^s could be expected, 

 and which were therefore simply as- 

 sumed to belong to the same type. 

 But of course the number of ears had 

 been difi'erent in different groups, 

 some being numerous and some rare. 

 Among the latter some had been mei 

 with in only a single head. As a 

 record has always been kept of the 

 exact number of heads used, it could 

 be made out to how many original 

 heads pure cultures were attributable. 

 It proved that the exceptional lots 

 were descendants of single heads in 

 each case. 



This almost accidental discovery, i;i 

 the summer of 1892, suggested the way 

 to a new method ot experiment, when 



the capabilities of the older one 

 seemed almost exhausted. In the har- 

 vest of that year, search was made for 

 new starting points. This time single 

 heads— in all about 2000 different heads 

 and panicles, representing as many 

 divergent types as possible, were 

 selected. 



The results of this trial exceeded all 

 expectation. For example, among 422 

 cultures of oats, ;{97 were pure, only 

 2.5 multiple. The exceptions were 

 thought to represent hybridizations; 

 and of course this possibility in all 

 the cases made one further selection, 

 of separate heads in each case, neces- 

 sary. But the high degree of unifor- 

 mity in the cultures of 1893 advanced 

 the principle ot the selection of single 

 individuals as a reliable source of 

 purity to the rank of an experimental- 

 ly established fact. 



From this fact it could be further 

 deduced that repeated selection, such 

 as that required in the method of elite 

 races would be unnecessa.ry — hybridiz- 

 in.g effects aside. The uniformity was 

 in fact such as to make a further selec- 

 tion impossible The differences which 

 formerly existed within the cultures 

 were now found to exist only between 

 the different groups, which now could 

 serve as the material of selection. 



From the industrial side the prin- 

 ciple of the sulficiency of one original 

 selection has a high value, for when 

 once the desired kind of purified grain 

 has been adopted, no further selection 

 of seed is required to keep the race up 

 to standard of productiveness, as in 

 the older method 



But purity is not all that is sought. 

 Excellence is the object of the cul- 

 turist's search. The question arose 

 whether the separate eulture.s would 

 possess sufficient variety of qualities 

 and excellences to afford the basis of 

 fuither advances. The results of the 

 sowings of 1S93 showed that their 

 mutual dilferences were much greater 

 than could be surmised from an inspec- 

 tion ot the original confused stock. 

 The separate cultures compiled with 

 the most diverse requirements, some 

 being highly resistant to frost, others 

 to disease, some being suited to light, 

 other to hard soils, some being early, 

 other late in rii)ening, some excelling 



QUEEN LOUISE 



CARNATIONS 



Choice field-grown plants. Price $4.00 per 

 100 ; S.is.oo per 1000. Large stock, and a bargain 

 at the price. Orders tilled day received, 



A. B DAVIS & SON, Purcellville, Va. 



AUGUST RdLKER & SONS 



Supply the trade wilh 



PLANTS AND BULBS 



For greenhouse or outdoor use, imported 

 to order. Send for estimate. Address — 



31 Barclay St^o^^P^^Po^ 752, New York. 



LANT5 



STAKES 



