Hereditas 40: 453-458 (1954)



FOR several years investigations on hybrids between wheat and rye have been made at the Institute of Genetics at Lund (cf. MUNTZING, 1939, 1943, 1948). In this connection a paper by PISSAREV and VINOGRADOVA (1944), dealing with the effect of embryo transplantation on the crossability of wheat and rye, is of special interest.

During their hybridization work, PISSAREV and VINOGRADOVA observed that the embryos and endosperm began to develop after pollination of the wheat flowers by rye pollen, but that in most cases further growth of the organs soon ceased completely. They consider that this disturbance partly depends on biochemical differences between the components used in the crosses and that this barrier may be overcome by embryo transplantation. By this technique it should be possible to modify the components used, with increased ability of hybridization as a result. Their arguments are strengthened by results gained by SCHANDER on the part played by the endosperm in the development of the plant (SCHANDER, 1934).

PISSAREV has worked out a technique for transplantation, described as follows: "The embryo together with the scutellum was cut from the dry caryopsis with the help of a safety-razor blade; care was taken to cut as little endosperm as possible. The embryo was then transferred on to the 'endosperm wilding' whose embryo  had preliminarily been cut out. The graft embryo was fixed in place by means of a paste prepared from a flour of the same genus to which the 'endosperm wilding' belonged."

According to PISSAREV and VINOGRADOVA, the wheat plant growing from the rye endosperm did not differ in morphological respects from the controls, but on the other hand there was a chemical difference. It was also apparent that the wheat transplanted on to the rye endosperm had a higher degree of crossability with rye compared with ordinary wheat. Three different varieties of wheat were investigated, Lutescens 62, Hybrid 170 and Aurora. The results of the crosses are summarized in the table below.

Percentage of successful crosses.

Variety Control group Grafted group
Lutescens 62  4,3 25
Hybrid 170  3,4 11,5
Aurora  2,8 19,2

This surprising result led to the present investigation carried out with Swedish material.


A preliminary experiment was started in the spring of 1952 when, during the last week of April, 200 transplantations of wheat embryos on to rye endosperm were carried out. The spring wheat variety 'Ella' and the spring rye variety 'Od' were the material used. The methodical instructions given by PISSAREV and VINOGRADOVA for the transplantation of embryos were followed exactly. About an hour after the wheat embryo had been transferred to the rye endosperm, the 'graft‑kernel' was placed in sterilized soil in small pots. At the same time 200 Untreated wheat kernels were also planted as control material. When the plants had reached a height of about 15 cms, they were moved from the greenhouse into outdoor frames. At the end of June 1005 flowers were emasculated in the transplanted group, as well as 206 flowers in the control group; the ears were isolated in pergamyn bags. The emasculated wheat flowers were pollinated with rye pollen after a week. About six weeks later the kernels produced were harvested. In the transplanted group, 17 hybrid kernels had been formed by the 1005 flowers pollinated, while on the other hand no kernels had been formed by the 206 pollinated flowers in the control group. The considerable difference between the number of flowers treated in the two groups was due to a long period of bad weather that hindered work, and also to an irregular supply of rye pollen.

It should be mentioned that both this year's emasculating and pollination work and next year's, which will be described below, was carried out by experienced personnel accustomed to such work, but without knowledge of the aim of the experiment.

On the basis of the above figures it was obviously impossible to draw definite conclusions as to the effect of embryo transplantation on the crossahility of the two varieties used. There was an indication of a positive result, however, which stimulated to further experiments.

At the beginning of April, 1953, a new experiment was thus started, this time on a larger scale. 500 transplantations of wheat embryos on to rye endosperm were then made. This group will from now on be called the V/R group. In addition, the same amount of transplantations of wheat embryos were made on to another wheat endosperm of the same variety. The latter group formed the control material and is called V/V below. As in the previous year's experiment, the transplantations were carried out according to PISSAREV and VINOGRADOVA'S instructions, and Ella spring wheat and Od spring rye were used this time, too.

The graft‑kernels were planted in sterilized soil in small pots, placed in a greenhouse at a temperature of 20ˇ C. 500 untreated wheat kernels were now also planted to determine the vitality of the sowing, as well as 200 detached embryos. The purpose of doing this was both to decide whether the detached embryos could be developed without their endosperm, and to what extent the embryos made use of the endosperm on to which they had been transplanted.

After about three weeks the material was moved from the greenhouse and put in frames. It was placed in groups with the four above‑mentioned groups side by side. With the exception of the plants that had their endosperm removed, the plants showed normal speed of growth and there was not any noticeable difference between the groups. The group with the detached embryos was even at an early stage less developed and had a lower speed of growth compared with the three other groups. In the middle of June when the plants were ready to be emasculated, there were 276 plants with ears in the V/R group and 417 plants with ears in the V/V group. 438 plants with ears had grown from the untreated wheat kernels and 27 from the detached embryos. 87 % of the untreated kernels had thus produced plants with ears, 83 and 55 being the corresponding percentages for the V/V and V/R  groups. Only 13 % of the detached embryos had germinated and grown into mature plants. It can at once be seen from the figures that there is a highly pronounced difference in the ability of the embryos to be developed if they have been detached, or if there has been an available endosperm they could utilize. The embryos transplanted to another wheat endosperm develop six times more often into plants than the detached embryos. The wheat embryos transferred to rye endosperm grow into plants with ears four times more frequently than the wheat embryos detached from their endosperm.

There is also a significant difference between the V/V and V/R categories. The percentage values 83 and 55 were obtained from the absolute ratios 417:83 and 276:224. A x2 test was made giving a x2 of 93 and a P smaller than 0,001.


2813 flowers from 101 plants in the V/V group and 2897 flowers from 102 plants in the V/R group were emasculated during the time from the 18th to the 20th of June. The ears were isolated after emasculation in pergamyn bags. After about a week the emasculated flowers were pollinated with rye pollen and the ears again isolated.

About the end of July and the beginning of August the experimental plots were harvested and the amount of kernels obtained after pollination was counted for each ear. It turned out that there were in all 75 hybrid kernels in the V/V group, i. e. the control group, while there were 400 in the V/R group, i. e. the experimental group. The figures for the successful crosses were 2,6 % for the control material and 13,8 % for the V/R group. Thus, the wheat transplanted on to rye endosperm formed about five times as many hybrid kernels as the wheat transplanted on to wheat endosperm.

The significance of the difference in the amount of kernels obtained between the groups V/V and V/R can be verified according to the x2 method below:

  Number of unsuccessful crosses Number of successful crosses n
V/V  2738 75 2813
V/R  2497 400 2897
Total 5235 475 5710

A comparison between the above ratios gave a x2 value of 230 which for 1 degree of freedom corresponds to a P-value much smaller than 0,001. As the two groups had been treated in the same way and grown up under similar conditions, the conclusion must be drawn that the factor that was different in the groups also gave rise to the difference in crossability. This factor has been the endosperm on which the respective embryos have been developed.

The results of the experiment described up to now can be summarized in the following table:

Group: V/V V/R Untreated wheat Detached embryos
Number of kernels planted 500 500 500 200
Number of plants with ears 417 276 438 27
Percentage of plants with ears 83 55 87 13
Number of pollinated flowers 2813 2897 -
Number of hybrid kernels 75 400 - -
Percentage of hybrid kernels 2,6 13,8 -

How the hybrid kernels were distributed among the different plants is apparent from the columns below:

of kernels
Number of plants Number
of kernels
Number of plants
  V/V V/R   V/V V/R
0 60 28 12 0 3
1 18 18 13 0 1
2 11 14 14 0 1
3 9 10 15 0 3
4 3 3 16 0  
5 0 5 17 0 0
6 0 2 18 0 1
7 0 0 19 0 1
8 0 4 20 0 0
9 0 3 21 0 0
10 0 1 22 0 1
11 0 2      

Evidently 60 plants in the V/V group were completely without hybrid kernels, while only 28 plants in the V/R group had not formed any kernels. The maximum formed has been 4 kernels per ear in the control material, but an ear with not less than 22 hybrid kernels was found in the V/R group.

From what has been said above, it is clear that part of the result that PISSAREV and VINOGRADOVA reported in their work quoted above could be reproduced. It has been shown with great statistical certainty that the crossability between Ella spring wheat and Od spring rye can be increased if the wheat plants used for the crosses are derived from embryos which were transplanted on to rye endosperm.

Also this year (1954) similar experiments are under way with kernels obtained from plants used in 1953 as material. Last year's hybrid kernels have been planted; the plants obtained from them will be studied.

Biochemical analyses of the material used is also being carried out parallel with this work. This is primarily aiming at a comparison of the composition of the aminoacids and carbohydrates in the different cereals used. Varieties of wheat and rye, other than those mentioned above, will also be studied.

June, 1954.


  1. MUNTZING, A. 1939. Studies on the properties and the ways of production of rye-wheat amphidiploids. Hereditas XXV, pp. 387‑430.
  2. —— 1943. N‡gra fšrsšksresultat med ragvete och tetraploidt horn. Nordisk Jordbrugsforskning. Hefte 5‑6, pp. 250‑262.
  3. —— 1948. N‡gra data fran forŠdlingsarbetet med tetraploid rag och ragvete. Nordisk Jordbrugsforskning. Hefte 1‑3, pp. 493‑5O7.
  4. PISSAREV, W. E. and VINOGRADOVA, N. M. 1944. Hybrids between wheat and Elymus. Comptes Rendus (Doklady) de l'Acad. des Sciences de l'URSS. Vol. XLV, No. 3.
  5. SCHANDER, H. 1934. Keimungsphysiologische Studien über die Bedeutung der Aleuronsehicht bei Oryza und anderen Gramineen. ‑ Zschr. f. Botanik 27, pp. 433‑515.