Gardening 6: 282-283 (June 1, 1898)

ADHERENCE TO TYPE IN SEED BREEDING
Will W. Tracy

The highest quality and most practically valuable seed of a variety is not that which from a large planting will produce a few superlative plants but that from which the largest proportion of the entire product will be of fine type. Uniformity of product is really the most important object of the seed breeding, and to secure it we must have, first, a clear conception of just what in every particular an ideal plant of the variety should be, and secondly, we must adhere rigidly to that ideal in our selection of breeders from year to year, and that we may do so it is essential that we have on record a full, minute and accurate description of the ideal plant we are working to produce, and select such plants, and such only, for breeders, rejecting those that differ from the ideal in any way, even if the difference is of itself an improvement.

Every living organism is the result of the balanced sum of hereditary influence inherited from its ancestors, that of one in a certain direction being modified by that of another in a slightly different or possibly directly opposite one. So that the more nearly alike all these influences are, the more certain and easily predicted is the result. If we introduce into the sum of heredity, through one or more ancestors, a more or less prepotent tendency in a certain direction, we disturb the balance and necessitate a new adjustment, and it is not certain that the resulting plant will be modified in the direction of the new influence in just the proportion that its numerical value bears to the whole, and in a majority of cases it will be found that some of the fruit will be greatly modified, others not at all, and still others will be changed in various directions, some possibly in one directly opposite to that of the new factor.

To illustrate, suppose we have a variety of corn which is generally ten rowed but occasionally showing an eight rowed ear and which would be more desirable if the ears were somewhat larger. We find a plant with a twelve rowed ear, larger, but in other respects like the type, and are tempted to use it as a breeder, hoping that the new twelve rowed tendency will overbalance that to produce eight rowed ears so that we would get rid of them entirely and increase the size of the ten rowed cars. It is possible that such might be the result, but I know by costly experience that there is greater probability that the outcome would be an occasional fourteen rowed ear, some twelve, some ten and more eight rowed ears than ever before; and while some of the ears might be larger, many would be undersized, and we should find all sorts of variation from the type, so that our product would be far less uniform and desirable than before.

My success in seed breeding has always been in direct proportion to the clearness of my conception of the ideal I was striving to produce and the persistency with which I adhered to that ideal in my annual selection of breeding stock. Over twenty-five years ago I had a five-acre field of very fine corn in which I spent several hours in studying the variety and forming an idea of what a perfect plant of the sort should be, how tall, how many and how broad the leaves, color of the silk and tassel, length of husk, length and character of the ears, character of the grain, etc., etc. I then wrote a minute description of this ideal plant, with the limit of variation in each particular admissible in a breeding plant. I then went into the field and it took me many hours to find a hundred plants which did not vary beyond the limit in one or more particulars. I selected the best ten as breeders, and each succeeding year, with my description of the ideal plant in hand, made similar selections. The result of the work was that the sixth year I had on that same five-acre field a crop over 50% of which was within the limits of variation established for breeding stock six years before, when not one plant in a thousand came within the limit.

I am certain that my success in thus fixing the desirable qualities of the variety came from the persistent adherence to the clearly defined type. It may be argued that this course leaves no chance for improvement, and it does not as to type, nor should there be any. When we "improve" the type, we change it and in just so far are establishing a different variety, for by variety as used in relation to plants propagated by seed we mean all those plants which are of a certain type, and the very idea necessitates that type being a fixed one; but there is not a stock of any variety, either of vegetable or flower seed in existence in which there is not room for improvement in the proportion of the product which will come absolutely true to the type of the sort. An intimate acquaintance with most of the stocks of vegetable and flower seeds in common use convinces me that the greatest horticultural need of the age is clearly defined ideas of just what ideal plants of the different varieties propagated by seed should be, and a closer adherence by seed growers to such ideals in selecting seed stock. It seems to me that the defining and describing such ideal plants is work which can best be done by our national Department of Agriculture working in connection with our state experiment stations Such descriptions, if well done, would be accepted by all seed growers and the result would be the weeding out of a great many sub-varieties and a vast improvement in the quality and practical value of our common varieties because of their closer adherence to universally accepted types.