September, 1907. 



f.».3 



American ^ee Journal 



)>=^g^^ J 



Breeding Bees by Selection 



for Improving their 



Honey-Producing 



Character 



uv r. w. h ^^I.AD^;N, v. k. «. 



It is ii fact well known lu lireedera of 

 •■ininials tliat to obtain uuil- maintain 

 ]ierniiini'nt iiii|irovoniont in auv character 

 it is necessary to breed stnciiy by selec- 

 tion from both parents in successive gen- 

 erations. Such breeding by selection has 

 iu many cases resulted in groat improve- 

 ments and in the production of many 

 new varieties. Domestic poultry are a 

 familiar example. In the case of bees 

 selection of the male parents has hith- 

 erto been practically impossible, at least 

 in this country, in all parts of which 

 ordinary bees are to be found, because 

 pairing takes place upon the wing some 

 distance from the hives, and we cannot 

 prevent some of our queens from pairing 

 with drones from neighboring apiaries. 



In my early efforts to improve the 

 honey-producing character of my bees I 

 found no great difficulty in selecting the 

 best honey-producing colonies, and breed- 

 ing from them in successive generations 

 on the female side; but on the male side 

 I had to content myself with rearing 

 each season a large number of drones of 

 best honey-producing parentage and 

 grandparentage, and eliminating as 

 nearly as possible all drones of other 

 parentage. By this means a considerable 

 number of unions between queens and 

 drones of best honey-producing parentage 

 were effected, but I found very great 

 difficulty in identifying the offspring of 

 these unions, so that in breeding the next 

 generation it was impossible in every 

 case to tell whether outside blood had 

 been included or not, and thus much of 

 the improvement that had been effected 

 was liable to be lost. 



In 1905, however, the great difficulty in 

 identifying the oft'spring of the unions 

 between queens and drones of my best 

 honey-producing parentage was practi- 

 cally overcome by the employment of a 



method of selection by color 



The object of the present paper is to ex- 

 plain more fully this method of selection 

 by color for the improvement of the 

 honey-producing character. 



The common bee of this country, as 

 everybody knows, has the ground color of 

 the body entirely black. In 1901 and 

 1902 I crossed my bees with selected in- 

 dividuals of one of the best honey-pro- 

 ducing strains of the American Golden 

 Italian bee, in which a considerable por- 

 tion of the abdomen is of a bright golden- 

 yellow color. The result in the second 

 and third generations of the cross-breds 

 was that the queens were exceedingly pro- 

 lific (in spring), and the bees very vig- 

 orous, but in almost every other charac- 

 ter these cross-breds were extremely var- 

 iable. Taking the two characters of 

 color and honey-production, I got — 



1. Dark-colored bees that were poor 

 honey-producers ; 



2. Dark-colored bees that were good 

 honey-producers ; 



3. Bright-colored bees that were poor 

 honey-producers; and 



4. Bright-colored bees that were good 

 honey-producers. 



as well as bees that were iutermeiliato 

 between these varieties. The Iji'sI, (colo- 

 nies of good honey-producers produced 

 larger amounts of honey than colonies of 

 onlinary Knglish bees, a fact which 1 

 attributed jjartly to the greater prolific- 

 iiess of tlie (lueens, jtartly to increased 

 vigor, :.nil partly to slight augmentation 

 iif till' good honey-producing character 

 due to cross-breeding. Some of the best 

 honey-producing colonies were of the 

 brightest color. It was therefore clear 

 that there was little or no correlation be- 

 tween the color character and the honey- 

 producing character — in other words, that 

 the two characters were inherited inde- 

 pendently. 



My next step was to eliminate all the 

 bright-colored bees except those that were 

 best honey-jiroduccrs, and to rear ()ueens 

 and drones from the latter only, and 

 these in the largest possible numbers. 

 Thus the only drones of bright-colored 

 parentage that paired with the queens 

 reared were of best honey-producing par- 

 entage. These drones considerably bright- 

 ened the color of their young, and thus 

 it became possible to distinguish, by the 

 brighter color of their young, the queens 

 that had been fertilized by drones of my 

 best honey-proUucing parentage from 

 those that had been fertilized by drones 

 of variable honey-producing parentage 

 and by drones from neighboring apiaries, 

 all of which produced darker young. This 

 distinction could be made as soon as the 

 first few hundred workers had hatched — 

 namely, iu less than a month after the 

 queen was fertilized. 



This method of selection by color for 

 the improvement of the honey-producing 

 character was also employed and ren- 

 dered more precise during the season of 

 1906. 



The laws governing the inheritance of 

 characters in cross-bred plants and ani- 

 mals have always been a puzzle to breed- 

 ers. Lately, however, they have been 

 carefully studied and much elucidated by 

 Bateson and others by means of elab- 

 orate breeding experiments. An excel- 

 lent account of some of these experi- 

 ments, with the conclusions they point 

 to, is given in a report to the Royal 

 Society by Prof. Bateson and Miss E. 

 R. Saunders, published in 1902. This im- 

 portant work confirms a remarkable law 

 which was first discovered and enunci- 

 ated by Mendel as long ago as 1865,- as 

 the result of experiments he made in 

 cross-breeding varieties of the garden 

 pea. Mendel 's discovery is too lengthy 

 to be given here, but the essential part 

 of it is " the evidence that the germ 

 cells of gametes produced by cross-bred 

 organisms may in respect of given char- 

 acters be of the pure parental types, and 

 consequently incapable of transmitting 

 the opposite character; that when such 

 pure similar gametes of opposite sexes 

 are united together in fertilization, the 

 individuals so formed and their posterity 

 are free from all taint of the cross. ' ' 

 (From Bateson and Saunders' Report, 

 page 12.) 



For instance, in the second and later 

 generations of cross-breds between hoary- 

 leaved and glabrous-leaved varieties of 

 the garden stock, Miss Saunders obtained 

 certain numbers of hoary-leaved individ- 

 uals which, when bred together, were 

 found to be incapable of producing glab- 

 rous-leaved individuals, and nearlv all 



the glabrous-leaved individuals that were 

 obtained, when bred together, were found 

 to be incapable of producing hoary- 

 leaved individuals. Atropa (the I)ea<lly 

 Nightshade) wag also found by -Miss 

 Saunders to obey Mendel's Jjiw in the 

 inheritance of yellow-colored and bbiek- 

 colored fruits. Prof. Bate.son found that 

 poultry obeyed Mendel's Law in the in- 

 heritance of "single" combs and 

 "rose" combs, and in that of "single" 

 combs and "pea" combs. More recently 

 Hurst has found that poultry obey Men- 

 del's Law in the inheritance of many 

 other pairs of opposite characters. In 

 their Report Bateson and Saunders give 

 a long list of pairs of characters in ani- 

 mals and plants that had, up to 1902, 

 been observed to obey Mendel 's Law ; 

 one of the most remarkable of these is 

 the waltzing habit and the absence of 

 the waltzing habit in mice. 



Bateson and Saunders ' work, which 

 only became known to me in the winter 

 of 1905-6, is likely to be a great help 

 in the work of endeavoring, in my cross- 

 bred bees, to obtain individuals with the 

 bright-color character and best honey- 

 producing character that are incapable of 

 transmitting auy mixture of the opposite 

 dark-color and poor honey-producing 

 characters to their young. In the case 

 of the bright-color character it is likely 

 that this object has already Deen attained 

 in some examples. The best honey-pro- 

 ducing character is a complex one, and 

 there are indications that it is largely 

 the result of the combination of several 

 characters, and that some of these are 

 more or less correlated to one another, 

 and that others, like vigor, are the direct 

 result of cross-breeding. The best honey- 

 producing character is partly dependent 

 on longevity (of workers), proUticness in 

 late spring (which acts favorably), pro 

 lifieness at other seasons of the year 

 (which may act unfavorably), and on 

 other characters. It is impossible to 

 forecast how much improvement in the 

 honey-producing character will be ob 

 tained when, by cross-breeding and selec- 

 tion, its component parts have to some 

 extent been separated and re-united in 

 new combinations. 



The honey-producing character of a 

 certain race or variety of bee may cer- 

 tainly vary very much in different cli- 

 mates. In the climate of some parts of 

 the United States the honey-producing 

 character of the Italian bee is exceed- 

 ingly good, but in the climate of Great 

 Britain, which during the honey-flow is 

 much colder, more cloudy, and more 

 windy than that of the United States, 

 the Italian bee is not a good honey-pro- 

 ducer, and the English bee is a better 

 one. This shows that breeding for the 

 improvement of the honey-producing 

 character in Britain must be carried on 

 in the British or in a similar climate; 

 aho that the English bee is a better bee 

 to work upon than the Italian. Crossing 

 the English bee with sufficient Italian 

 blood to enable one to brighten the color 

 for practicing selection by color improves 

 it for the purpose of breeding for the 

 improvement of the honey-producing 

 character, because it gives it increased 

 prolifieness in spring, vigor and vari- 

 ability. 



In the second and third generations of 

 my cross-breds it was a common thing 

 to find a queen that would produce some 



