Proc. Nat. Acad. Sci. 1918; 4: 246-25
Communicated by T. B. Osborne, July 12, 1918

The experiments dealing with the effects of inbreeding and crossbreeding in the naturally cross-pollinated corn plant, Zea mays L., carried on at the Connecticut Experiment Station have been reported on previously by East2 and Hayes.3 A continuation of these experiments shows that there is no appreciable reduction in vigor and productiveness, as indicated by the yield of grain and height of plant, after the eighth generation of self-fertilization as can be seen in table 1. The resulting inbred strains, now in the eleventh generation of continuous self-fertilization, are remarkably uniform and constant. Very little change in average ear row number or reduction in variability of row number has taken place after the eighth generation in two inbred lines derived originally from the same plant in the second generation of inbreeding as shown in figure 1. These results are in agreement with the theoretical attainment of nearly complete homozygosis automatically in about the eighth generation of self-fertilization. Although the inbred strains are now only about one half as productive as the original variety, the plants, judging from their behavior in the past three years, are capable of being maintained unchanged indefinitely by self-fertilization.




1-6-1-3-etc. 1-7-1-1-etc. 1-7-1-2-etc. 1-9-1-2-etc.
per acre
height Yield
per acre
Height Yield
per acre
Height Yield
per acre
    bushels inches bushels inches bushels inches bushels inches
1916 0 74.7 117.3 74.7   117.3 74.7   117.3 74.7   117.3
1905 0 88.0   88.0     88.0     88.0    
1906 1 59.1   60.9     60.9     42.3    
1908 2 95.2   (1907) 59.3     (1907) 59.3     51.7    
1909 3 57.9   (1908) 46.0     (1908) 59.7     35.4    
1910 4 80.0   63.2     68.1     47.7    
1911 5 27.7 86.7 25.4   81.1 41.3   90.5 26.0   76.5
1912 6             (1913) 38.9    
1913 7 41.8   39.4         (1914) 45.4   85.0
1914 8 78.8 96.0 47.2   83.5 58.5   88.0 (1915) 21.6    
1915 9 25.5   24.8         (1916) 30.6   78.7
1916 10 32.8 97.7 32.7   84.9 19.2   86.9 (1917) 31.8   82.4
1917 11 46.2 103.7 42.3   78.6 37.6   83.8    

FIG. 1.

In contrast to the uniformity of all the plants within one inbred line is the difference between the several lines. With regard to ear characters two of them have fiat cobs and two have round. One has uncolored cobs while the others have colored. They all differ in the shape and size of ears and seeds and in the arrangement of the seeds on the ears. Equally noticeable differences characterize the tassels, stalks and leaves.

One of the most pronounced effects of inbreeding in maize is the appearance of sterility in the form of pollen and ovule abortion. An extreme reduction in the amount of pollen produced is shown frequently in plants long inbred. Not all strains show this, however, as some have well developed anthers and produce abundant pollen. In nearly every case, however, those inbred strains which have the best developed staminate inflorescences have poorly developed pistillate inflorescences and those strains which have the largest pistillate inflorescences and are most productive of grain produce a very small quantity of pollen. Doubtless many plants are produced which completely lack one or the other or both functions but such types, of course, can not be maintained in self-fertilization. There is, therefore, a decided tendency for inbreeding to change a monoecious plant into a functionally dioecious plant.

All of the inbred strains obtained show an immediate return to the vigorous condition of the non-inbred parents when crossed. The return of vigor is greater in some combinations than in others. In general there seems to be a correlation between the productiveness of the female parent strain and the productiveness of the first generation hybrid. This is in part due to the fact that hybrid plants start from better developed seeds when the female parent is the more vigorous. It is also shown in a comparison of the F2 generation, starting from large seeds grown on vigorous hybrid plants, with the F1 generation starting from seeds produced on the reduced inbred plants. The first generation overcomes this handicap and in every case may be expected to surpass the second or subsequent generations as shown by the curves in figure 2.

The manifestation of heterosis is greater in some characters than in others. In the many crosses between inbred strains observed yield of grain is increased 180%, height of plant 27%, length of ear 29%, number of nodes 6% and the number of rows of grain on the ear 5%. Abundant evidence has been obtained from these inbred strains to show that the endosperm also is immediately affected by crossing. Differences in weight of reciprocally crossed seeds as compared to selfed seeds developed in the same inflorescences have averaged from 15 to 20% in favor of the cross. Furthermore in spite of this increase in weight the hybrid seeds mature faster as shown by their lower water content. Due in part to this latter fact there is a noticeable difference in viability as shown by the per cent of germination between the selfed and crossed seeds produced by the same plants. The crossed seedlings also appear from one to two days earlier than the selfed seedlings.

FIG. 2.

In view of the notable advantages to be gained from cross-fertilization it was thought that when mixtures of a plant's own pollen with that of a distinct sort were applied there might be a selective fertilization favoring the foreign pollen. By taking advantage of xenia it was possible to obtain and to distinguish selfed and reciprocally crossed seeds from plants of two different inbred strains. Since the same mixture of pollen was applied to plants of both strains, the number of seeds produced by one kind of pollen should be in the same ratio to the number of seeds produced by the other kind of pollen, on both types of plants to which the mixture was applied, irrespective of the amounts or viability of the two kinds of pollen used. Any deviation from a perfect proportion in the form of an excess of crossed seeds or an excess of selfed seeds would indicate selective fertilization in one or the other direction. A large amount of data has been obtained in this way which shows that there is no inequality in fertilizing ability favoring foreign pollen. In fact the data show a selection in the other direction large enough to have significance when compared to the probable error. Whether this is really the case or whether there has been a constant error in classifying the seeds will have to await the growing of the plants from these seeds in order to determine the accuracy of the classification. In the meantime it hardly seems possible that the error could be so great as to hide an actual selective fertilization in favor of cross-pollination, since the inbred strains are so uniform in color of endosperm and the crossed seeds usually so distinct that the error of classification is small and presumably not all in one direction. And if there is no selection in favor of cross-fertilization, there is then no effect of crossing until the zygote is formed however great the advantages immediately accruing to the resulting embryos and endosperms.

  1. An extract from a treatise submitted to the faculty of the Bussey Institution of Harvard University in partial fulfillment of the requirements for the degree of Doctor of Science, December, 1917, and to be published in full as a contribution from the Connecticut Agricultural Experiment Station.
  2. East, E. M., Connecticut Agric. Exp. Sta. Rep., 1907, 1908, (419-428); Amer. Nat., Lancaster, Pa., 43, 1909, (173-181).
  3. East, E. M., and Hayes, H. K., Washington, U. S. Dept. Agric., Bur. Plant Ind. Bull. No. 243, 1912.