EXPERIMENTS IN BREEDING SWEET CORN. 283 



The same thing is brought out in a still more striking way so 

 far as yield is concerned by comparing ear No. 578 with the 3 

 daughter ears from No. 50. Ear No. 578 is a daughter ear 

 from the very poorly shaped, scrubby ear No. 22 shown in Fig. 

 227. Now while 578 is lacking in some particulars the high 

 quality of Nos. 574, 576 and 577 yet after all it is a very good 

 ear — far better than No, 22. But if we turn to the rate of yield 

 of shelled corn to the acre it appears that the row from ear No. 

 578 (the grand-daughter row from ear No. 22) yielded nearly 

 10 per cent, better than any grand-daughter row from ear No. 

 50. In other words, suppose a farmer had been selecting ears 

 for seed ; the facts show that the outcome, so far as concerns 

 the number of pounds of corn to be hauled to the factory, would 

 have been better if ear No. 22 had been used to* found a strain 

 of seed rather than ear No. 50. Yet no one would ever think 

 of using such an ear as No. 22 for planting if he could get any- 

 thing better. 



Such results as are here being discussed should not under any 

 circumstances be taken to mean that the right thing to do when 

 selecting seed is to pick out ears like No. 22, rather than those 

 like No. 50. What the results do mean is that the external 

 qualities of the ear, whether good or bad, are a very poor indi- 

 cation what that ear will do when planted We have here simply 

 another illustration of the old adage to the effect that it is not 

 possible to tell how far a frog can jump by mere inspection of 

 the frog. His jumping capacity is determined by innate, invis- 

 ible qualities and characters, only to be tested by making him 

 actually jump. In precisely the same way it is not possible to 

 tell by the appearance of the ear how well relatively it is going 

 to yield when planted. Because, just as with the frog, the yield- 

 ing capacity depends on innate, invisible qualities. The only 

 way to tell how it will yield is to plant it and see. If it then 

 does yield well it zvas a good ear, even though it may not have 

 looked the part. 



All this means practically that in selecting corn for seed the 

 selection must be on tJie basis of the perforniancc of the proge)iy 

 (here the row) as distinct from individual ear selection. The 

 best ears to select for seed are those which came from good 

 roivs. A poor ear from a good row is vastly better than good 

 ears from a poor row when the planting is on the ear-to-row 



