522 SECTIONAL TRANSACTIONS.—M*. 
PT gives red spines and red fruits, 7 tinged spines and red fruits, P green spines and 
apricot fruits, and pt green spines and yellow fruits. 
The origin of many of the polyploid fruits is involved and their genetic constitution 
is complex. Consequently, for whatever purpose they are raised, larger families are 
required in such fruits than in diploid forms. 
5. Cytological studies. These investigations have shown that the principal 
varieties of our cultivated fruits are diploids or even-numbered polyploids, the latter 
in all probability of hybrid origin, for this appears to be a condition of their fertility. 
Where odd-numbered polyploids such as triploids and pentaploids are found, they 
are generally less fertile and of comparatively little value. 
The condition of the apple group is exceptional ; seed fertility is not so closely 
related to fruit-production, and ‘ triploids ’ are consequently of economic importance. 
The ‘diploid’ apples are, however, themselves of exceptional constitution; their 
gametic number of chromosomes is seventeen, but my colleagues C. D. Darlington 
and A. A. Moffett have concluded from the type of chromosome association they have 
observed that this number is the result of reduplication of an original set of seven in 
different proportions. 
The results of these combined genetic and cytological studies have shown that 
polyploidy and hybridisation have played an important and progressive part in the 
evolution of many of our important fruits. The studies have also afforded information 
regarding taxonomic relationships and the practical possibilities of cross-breeding 
both with species and varieties. a 
Mr. W. J. C. Lawrence.—The Genetics of Dahlia and its Bearing on - 
Inheritance in Cultivated Plants. 
The garden dahlia (Dahlia variabilis) was introduced to Europe in 1789. Its 
. subsequent rapid variation under cultivation is probably without parallel in regard 
to the diversity of colour and form produced. 
As the result of morphological, genetical and cytological observations it has been 
possible to explain the reason of this wide variation and to construct a scheme showing 
the probable origin and constitution of D. variabilis. The value of these experiments 
is that they contribute to our knowledge of many important plants of similar constitu- 
tion to D. variabilis, but not so amenable to experimental requirements. 
With the exception of D. variabilis the species of Dahlia may be classified in two 
groups for flower colour: (i) ivory or magenta, (ii) yellow, orange or scarlet. 
D. variabilis unites both colour groups within itself. 
Flower colour in Dahlia is the expression of two series of soluble pigments : 
(i) flavones, (ii) anthocyanins. The flavones form the ground colours (ivory and 
yellow) on which the anthocyanins are superposed. Pale anthocyanin on ivory gives 
magenta flower colour ; on yellow, apricot. Deep anthocyanin on ivory gives purple ; 
on yellow, scarlet. 
The results of the breeding experiments are as follows: D. variabilis is self- 
incompatible, setting no seeds with its own pollen. Two independent factors, I 
and Y, govern the production of ivory and yellow flavone colours respectively. The 
inheritance of Y is tetrasomic, giving the characteristic ratios 5: 1, 11:1 and 35: 1. 
The inheritance of J is tetrasomic also, but dominance is incomplete. 
Anthocyanin production is also governed by two independent factors, A and B. 
B produces deeper pigmentation than A. The inheritance of A and B is tetrasomic. 
A is cumulative in action. 
Another factor or factors inhibit, cumulatively, the formation of yellow flavone, 
thus giving rise to the intermediate ground colours cream and primrose. Anthocyanin 
on these grounds gives intermediate flower colours such as crimson. The inheritance 
of the inhibitor is tetrasomic. 
The profuse variation found in D. variabilis, therefore, is mainly due to :— 
1. Self-incompatibility, which enforces cross-pollination, and thus maintains the 
heterozygosity of the species. 
2. The occurrence of tetrasomic factors, giving a very large number of genotypic 
combinations. 
3. The cumulative action of some of these factors, which further adds to the 
diversity of form. 
Cytological observations indicate that D. variabilis is an octoploid species (2n=64). 
Five other species examined were found to be tetraploids (2n=32). 
