Breeders who are conducting careful experi ments will find it necessary and desirable to use what may be termed statistical methods of judging their plants. While we are breeding possibly for one primary improvement, as, for example, in-' creased yield, it is necessary, at the same time, that we should keep the product up to the standard in other characteristics, namely, quality, disease resistance, drought-resistance and the like, and that we see that all of the good qualities of the variety are retained. To do this properly necessi tates the use of a score-card, on which each char acter of the plant which is im portant is given its relative weight or grade. By the use of such a score-card the breeder can judge each character separately, and by the adding up of the scor ing get the rank of different plants in a comparative way (Fig. 83).
Test of transmitting power.
A factor of primary impor tance in all breeding work is the testing of what is termed the transmitting or centgener power. It is necessary for us to know that a certain plant, which, for example, gives a heavy yield, has the faculty of transmitting this tendency of producing heavy yield to its progeny (Fig. 84). It is frequently found that two select plants that are equally good so far as their yield is con cerned will give progeny which, as a whole, differ greatly in this respect. In the progeny of one almost every plant may have in herited the desired quality, while in the progeny of the other only a few of the plants may show in any noticeable degree the inheri tance of the quality. To determine the prepotency or transmitting power, it is necessary to grade carefully the progeny of each individual ; and this is the primary reason for planting the progeny of different individuals in separate rows or separate plats, so that they may be examined easily. (Fig. 85.) It would seem to be an easy matter, when we plant the progeny of different plants in rows or small plats by themselves, to get the comparative yield, for example, of 100 plants, and from this to figure up the average percentage of the transmitting or centgener power. This matter, however, is very difficult in many cases. In corn, for example, cer tain individuals may stool and form suckers that have fairly good-sized ears. If the cffrn is planted thin enough on the ground these suckers will tend to increase the yield, and render the proper judg ment of the transmitting power very difficult. It would seem at first thought that such suckering, if it increased the yield, would be desirable, and should be considered a favorable character in con nection with the individual. However, if the soil is heavy enough to have allowed this suckering to give increased yield, it would have been possible on the same soil to have placed the plants closer, and, as seed is of little comparative value, it would be best to have a non-suckering type, and plant the corn as closely as the soil would properly per mit. Again, it is almost impossible to get perfect stands, and a change in the stand may affect the yield. Very many difficulties and problems enter into the figuring out of this transmitting power, and it is obviously impossible to give directions for all cases. The breeder must study conditions
and determine carefully what policy to pursue in each case.
The use of hybridization in plant-breeding.
Ever since the time of Knight, hybridization has been used extensively by plant- breeders, and it seems that this is the only sure means of forcing variations. Whenever it is possible to secure dis tinct species and races that can be hybridized, it is possible greatly to increase the variation in differ ent directions, and thereby afford opportunity for greater selection than would otherwise be possible. Plant-breeders have come to understand that when desirable characters are exhibited by different species or races it is possible frequently, if not usually, to unite these characters in a hybrid if the work is done intelligently and on a large scale. (The writer uses the term hybrid here in a general sense, referring to any product of a cross when the parents were noticeably distinct from each other, whether the parents belong to different races, clone, varieties or species. It may be stated that this general or broad use of the term hybrid has become almost universal in recent years.) When plants of different races are crossed, as, for example, different races of wheat, corn or cotton, the hybrid usually comes nearly intermediate between the two parents in the first generation. And this is the case also when different fixed species are crossed. If, how ever, individuals belonging to unfixed races are crossed, there is usually a considerable variation in the first generation. This is well illustrated by the crossing of different clone of apples, pears, oranges, and the like, when the different so-called varieties are simply transplanted parts of the same individual seedling which have not been bred to a fixity of type. It is well known that if seeds of an apple variety be planted, the resulting plants exhibit many different variations in the first generation. The parents themselves, therefore, not being of fixed type, when they are hybridized they produce progeny which in the first generation is variable. An illus tration is afforded in the crosses made by the writer of the trifoliate orange with the ordinary sweet orange, in which the hybrids of the first generation vary in fruit, foliage and branching qualities, so that almost every individual differs markedly from every other individual of the same combination. In the crossing of races which have been bred true to type, whether of the same or of different species, the first-generation hybrids, however, are nearly uniform in the characters presented, and in such instances it is necessary to secure a second gen eration of the hybrids in order to accomplish the breaking up of the characters and the production of a large number of variations. Ordinarily, there fore, desirable variations are looked for in the second generation. This, as has been explained above, is true only in the case of hybrids of species and races that are fixed in type.