Hereditas 39: 236-240 (1953)


The problem of dominance is of secondary interest in genetics, since dominance is probably caused by different phenotypical factors, anatomical or chemical and physical. It is well known that the dominance of a given gene may change with changes in the external environment. But also changes in the genic environment, caused by the interactions of the genes, may result in changes in the expression and also the dominance of a given gene. Cases where a change in domin- ance is caused by the genic environment are, therefore, of a certain interest; two such cases are here reported. ERNST NILSSON (1929) has reported a similar case.

Change of dominance in a flower colour gene in Lupinus anqustifolius.—In a paper in volume II of Hereditas (HALLQVIST, 1921) the present author has reported the inheritance of flower colour in Lupinus angustifolius. The following four genes were found to influence the flower colour:

B: blueish red.
V: purple,
BV: blue,
R: pure red, without any blueish tint. This gene is also a fundamental colour gene, rr plants are white.
F: full colour intensity, ff plants are dilution forms of the colours determined by the other genes.

As is generally the case with albina types, the rr plants reported in the paper quoted were completely without anthocyanin in all vegetative parts, including the sepals. At first also the petals were pure white but on ageing they developed a colour which was always quite evident and sometimes fairly strong. This phenomenon is typical of the genus Lupinus where all flower colours become darker with age. The white plants of this type were not true albinas but might be classified as pseudo-albina.

In further experiments I found a type without age colouring, completely free from any trace of anthocyanin in all parts of the plant; a true albina. It occurred in the following combination:

Weak red colour ×
White with age colouring

and the true albina was probably constituted rrff.

Both R and F are dominant and the expected segregation in F2 should be: 9 strongly coloured: 3 weakly coloured: 3 white with age colouring: 1 pure white. Five F2 families gave the following result:

found 126 37 23 39  
expected 126,6 42,2 42,2 14,1  
X2 0,00 0,64 8,74 43,97 = 53,35
        P < 0,001

The agreement is very poor and there is evidently some disturbance. Linkage is excluded since it should have resulted in a surplus of parental types; instead, the double recessive occurs in excess, the double dominant in normal proportions.

The disturbance occurs only within the rr-group. According to earlier  results, the 62 rr plants should have given 3/4 pseudo-albina, i.e. 46,5 pseudo-albina and 15,5 true albina. The actual proportions are nearly reverse, i.e. 23 pseudo- and 39 true albina. This result may be fully explained by the hypothesis that the usual dominance in the F gene, F > f, has become reverse in the presence of homozygously recessive rr. If such is the case, the dihybrid segregation should be 9:3:1:3 in F2 or:

expected 126.6 42.2 14,1 42,2  
found 126 37 23 39  
X2 0,00 0,64 5,62 0,24 = 6,50
        P > 0,05

The agreement is fairly good with this hypothesis.

Change of dominance in chlorophyll genes in barley.— In earlier papers in volumes IV, V and VIII of Hereditas the author (HALLQVIST, 1923, 1924, 1926-27) has reported the occurrence and segregation of number of chlorophyll mutants in barley. One group of such mutants has been named virescens, in which the seedlings are at first yellowish white and later become light green in the tips of the leaves.

Diallel crosses have been made between three such types, viz. virescens I, II and III. In all three crosses the double recessive homozygote is completely devoid of chlorophyll; it is as pure white as the albina mutants. At the same time the leaf is as narrow as that of true mutant and the bifactorial combination of two virescens genes, thus, phenotypically completely imitates a monofactorial albina mutant.

In dihybrid segregations of two virescens it is often difficult to classify the two genes and both virescens are therefore usually classified together. In a normal, undisturbed segregation the expected proportions are, thus, 9 green: 6 virescens: 1 white.

In the cross virescens I x virescens III these proportions are realized.

  green virescens white  
found 3668 2364 387  
expected 3610,7 2407,2 401,2  
X2 0,91 0,78 0,50 =2,19
      P = 0,50-0,30

The agreement is very good.

When virescens II is included in the cross the results are quite different.

  green virescens white  
  Virescens I × virescens II  
found 144 72 46  
expected 147,4 98,3 16,4  
X2 0,08 7,04 53,42 = 60,54
      P < 0,001
  Virescens II × Virescens Ill  
found 12774 5629 4143  
expected 12682,1 8454,8 1409,1  
X2 0,67 944,5 5305 = 6250,17
      P < 0,001

In both cases the number of green plants is quite normal, while there are about 3 times as many white ones as expected, with a corresponding deficit in virescens.

As was the case in the lupine, linkage is excluded as an explanation. Most probably the phenotypically white group has a different genotypical composition from the one expected, and the most reasonable hypothesis is that recessive homozygosity in VII results in a change in dominance in the pairs VI-vI and VIII-vIII. Phenotypical albina should then occur not only in the two double recessives vIIvIIvIvI and vIIvIIvIIIvIII but also in the biotypes vIIvIIVIvI and vIIvIIVIIIvIII, and the segregation should be 9:4:3 instead of 9:6:1. According to this hypothesis, expected numbers and X2 in the two crosses should be

  green virescens white  
  Virescens I × virescens II
expected 147,4 65,6 49,2  
X2 0,08 0,99 0,21 = 1,18
  Virescens II × virescens III
expected 12681,1 5636,4 4227,3  
X2 0,67 0,01 1,68 = 2,36

In both cases the agreement is very good.

There is one further reason to accept the hypothesis. In the cross virescens II X virescens III the albina group is not quite uniform. There are numerous plants with the colour and narrow leaf of albina which have, however, a very minute green spot at the tip of the leaf. It is reasonable to suppose that these plants are the VIIIvIII plants which because of the change in dominance have become albina but with a very small effect of the VIII allele. If this is true, 2762 of the 4143 plants classified as albina should have shown this minute green spot; it was observed in only 2573 plants, however. The deficiency may be due to difficulties  in classification, since the very small green spot is rather difficult  to observe; it may also be due to a weak penetration of the single VIII allele.

There is a striking similarity between the two cases here reported. In both cases recessive homozygosity in one gene has resulted in a change in dominance in another pair of alleles. In lupine the constitution rr had the result that the Ff plants became phenotypically similar to ff plants. In barley the combination vIIvIIVIvI has exactly the same albina phenotype which else is typical of double recessives in virescens genes. In vIIvIIVIIIvIIIvIII the change towards albina is not complete, as the single VIII may be traced in most cases by a minute green spot, but the dominance for VIII has changed into a strong prevalence for the other allele, vIII.

It is interesting to note that in both cases the genes involved have reducing effect. Both r and f reduce the anthocyanin colouring in lupine; vI, vII and vIII all reduce the chlorophyll in the plastids in barley. The reducing effect of the genes r and respectively, may be especially strong, which might explain their effect in changing the dominance of other genes. When the two genes mentioned are homozygously recessive the colour is already so reduced that the change from normal to mutant of one single further allele in another gene sufficient to result in a complete absence of the colour. Usually and other genic environments a corresponding change of one allele of the latter genes has no outwardly visible effect.

The above recorded experiments were carried out as early as in 1925-1928 at the Weibullsholm Plant Breeding Institute, Landskrona.


  1. HALLQVIST, C. 1921. The inheritance of the flower colour and the seed Colour in Lupinus angustifolius. — Hereditas II: 299-363.
  2. —— 1923. Gametenelimination bei der Spaltung einer zwerghaften und chlorophylldefekten Gerstensippe. Hereditas IV: 191-205.
  3. — 1924. Chlorophyllmutanten bei Gerste. Ihre Entstehung und primaren Spaltungen. Hereditas V:49-83.
  4. — 1926-27. Koppelungen und synthetische Letalitat bei den Chlorophyllfaktoren der Gerste. Hereditas VIII: 229-254.
  5. NILSSON, E. 1929. Erblichkeitsversuche mit Pisum. I. Unterdrückung der Dominanz eines Faktors durch die Wirkung anderer genetischer Faktoren. — Hereditas XII: 17-32.