564 



AMERICAN BEE JOURNAL. 



tables that almost nothing has been 

 made of 3A and 2C. The explanation 

 of this is that the latter persisted in its 

 desire to swarm until it eventually lost 

 a considerable part of its bees by their 

 uniting with another swarm, and the 

 former, within a day or two after 

 swarming, in some way lost its queen 

 and dispersed more or less in conse- 

 quence. The only question with regard 

 to the propriety of this course arises 

 when we consider table C, wherein the 

 comparative advantages of large and 

 small swarms are weighed. Perhaps 2,C 

 should have been permitted to cut some 

 figure in that, for it clearly illustrates 

 one of the disadvantages of very large 

 swarms. 



Table B is a summary of table A and 

 puts the tables of each group of colonies 

 employed in the experiment side by side 

 so that the general results are seen at a 

 glance. 



Table C is derived from table A and 

 puts in constrast the vs^ork of the 

 stronger colonies of each group with 

 that of the weaker ones of the same 

 group, and table D is an epitome of 

 table C. 



Now what do the tables teach with re- 

 gard to the comparative profit of the use 

 of starters, foundation, and comb in the 

 brood-chamber as well as with regard to 

 the advantage of larger and smaller 

 swarms? It would be too much to ex- 

 pect that upon either of these points the 

 results shown by the several hives taken 

 separately would invariably point in the 

 same direction. There are so many in- 

 scrutable influences at work that we 

 may well look for unexplainable vagaries 

 in the revelations of individual hives. 

 It is largely for this reason that I think 

 the writer who in one of the apicultural 

 journals recently very flatteringly inti- 

 mated that the results obtained in the 

 experiments at this branch of the Michi- 

 gan experiment station would be conclu- 

 sive, was hasty. 



If several varieties of wheat, for 

 instance, were sown side by side upon 

 precisely the same kind of soil so far as 

 human skill could determine, and each 

 variety should receive exactly the same 

 treatment in all respects and at the 

 same time, and one certain variety was 

 found to yield 20 per cent, more than 

 any other, yet the farmer who should 

 from the one experiment jump to the 

 conclusion that the result would always 

 be the same would be accounted lacking 

 in judgment. The results must be veri- 

 fied repeatedly before they can be ac- 

 cepted as the rule. Just so it is with 

 the matter in hand. Yet the multiple 



character of our experiment with results 

 so nearly uniform give strong assurance 

 that what seems to be disclosed is in the 

 direction of the truth. 



From the figures given in the last col- 

 umn of table B, we find that the colonies 

 hived on comb gained in all more than 

 11 per cent, over those hived on starters, 

 and that those hived on foundation 

 gained more than 13 per cent, over the 

 same. But if we examine with refer- 

 ence to comb ho7iey only, we find that 

 colonies "A" (those on comb) gain less 

 than 5 per cent, more than colonies "C" 

 (those on starters), while colonies "B" 

 (those on foundation) gain more than 

 17 per cent, over " C." But it may be 

 said that " C " has an undue proportion 

 of the weaker colonies, which is true ; 

 still, if we turn to table "C" and con- 

 sider only the strong swarms in each 

 group, we find that "A" gains 93^ per 

 cent, more than "C" in comb honey, 

 and " B" gains 42 per cent, more than 

 " C !" But strange to say, taking the 

 light swarms in the same table and col- 

 umn the positions are exactly reversed — 

 " A" gains nearly 3^ of one per cent, 

 over " B," while "C" gains nearly 32 

 per cent, over "B." If space permitted 

 it would be interesting to inquire why 

 the diCference in the weight of the colo- 

 nies should cause this reversal in their 

 positions in regard to the amount of 

 comb honey produced. 



Referring again to table "B" from 

 the figures given in the third column, 

 where the gain for the first period is 

 given, we deduce that "B" gains dur- 

 ing that period more than 53 per cent, 

 over "C," while " A " gains more than 

 68 per cent, over "C ;" but during the 

 second period the figures show that for 

 that period the positions are exactly re- 

 versed : while for the third period the 

 positions as to relative gain are again 

 changed, "A" making a spurt and leav- 

 ing " B" in the rear. 



Referring again to table "C," we find 

 that the strong colonies invariably gain 

 the more in the first period, while the 

 light ones take a decided lead both in 

 the second and the third periods ; in the 

 amount of comb honey for the entire 

 time in each group, the strong colonies 

 have a decided advantage, and so in 

 groups " A " and " B " in amount of 

 total gain, but in group "C,"in this 

 point, the weaker ones are far in the 

 lead. But this sort of comparison might 

 be carried on almost endlessly. 



If we examine table " A " we find, as 

 was to be expected, that the results in 

 the cases of some individual colonies do 

 not always accord with the general re- 



