Odd multiple polyploids are relatively infertile, consequently in fruits such asRubus and.Prunus, where seed and fruit development are closely associated, triploid and other odd multiple forms are relatively unproductive. Triploid apples however are productive,e.g. Bramley’s Seedling, a triploid, is probably more widely cultivated in this country than any other apple.

In apples a very low proportion of fruit to flowers is sufficient to give a yield. The apple has ten embryos, and often a single seed is sufficient for the development of a fruit, and even this seed may be imperfect. This approaches parthenocarpy and renders fruit production still less dependent on the formation of seeds. Fruitfulness in apples may therefore be maintained in spite of a high degree of generational sterility.

Triploidy in apples is another example of the occurrence of sterile forms in species where a substitute has been found (either in nature or in cultivation) for normal seed and fruit production. The substitution in apples is more complex than usual for, while sexual reproduction is now replaced by grafting, the necessity for the stimulus of seed growth in the formation of a fruit is largely evaded. Therefore triploids are able to fruit although incapable as a rule of yielding offspring of any value.

The offspring of triploids, whether derived from selling or crossing with diploids, lack vigour, presumably owing to their aneuploid constitution (cf. Darlington and Moffett). Consequently triploid varieties are likely to be of little value in practical breeding as the necessary vigour and fertility would rarely be obtained in the resulting offspring.

The inheritance of a “purple anthocyanin” character in apples is described.Malus Neidzwetzkyana was found to differ fromM. Malus in having a single dominant gene for “purple pigmentation”. Great variation was found within the “purple” plants, probably due to the segregation of modifiers.

A technique for increasing the percentage of germination by removing the testa is described. The percentage of scorable plants was increased from 56 to 99. No differential viability of the “green” and “purple” embryos was found.

1. The diploid variety of pear “Fertility”, 2n = 34, gave rise by somatic mutation to a new large-fruited form. This new form was found to be tetraploid, 2n = 68.

2. The tetraploid form was found to be more self-fertile than the type.

3. Pollen formation was that expected in a non-hybrid tetraploid.

4. The results obtained from using the diploid form of “Fertility” as female in crosses with triploids indicate that in certain crosses apomixis and/or gametic selection occurs.

5. The diploid form is shown to be somatically unstable, though meiosis proceeds normally.

6. The value of tetraploidy in producing new polyploid forms of pears is briefly discussed.

1. The reproductive behaviour of a number of species and varieties ofRubus has been investigated.

2. Diploid species always behave sexually. Occasional unreduced germ-cells occur and take part in fertilization, giving rise to polyploid forms.

3. In polyploid forms and species reproduction may be entirely sexual, entirely non-sexual or partly sexual and partly non-sexual (apomictic).

4. Polyploid species vary in the degree to which apomixis is developed, and a particular species may show a variation in reproductive behaviour depending on the species used as male in cross-pollination.

5. Segregation has been found to occur within non-sexual offspring. Hence test crosses, in conjunction with cytological and embryological studies, are necessary to determine the precise mode of reproduction.

6. The results are discussed in relation to the difficult taxonomy ofRubus.

1. Although the loganberry arose as recently as 1881, the details of its origin are in dispute. It has been held (1) that the loganberry is a hybrid, the result of a cross between a blackberry and a raspberry, and (2) that it is not a hybrid but a “direct derivative” of the wild blackberry or dewberry of California,Rubus vitifolius. The present experiments were planned to decide between the two theories.

2. From crossingR. vitifolius, 8x, withR. idaeus, 4x, a hybrid was obtained which in morphology, chromosome number, sex and fertility closely match the loganberry; like the loganberry it is hexaploid and nearly true-breeding.

3. Crosses made betweenR. vitifolius, R. idaeus, 2x, and the loganberry, in whichR. vitifolius andR. idaeus chromosomes were brought together in different proportions, gave results which are also in agreement with the hybrid view.

4. The only conclusion which can be reached from the experiments is that the loganberry is a hybrid as originally supposed, derived from an unreduced male germ cell of a raspberry, which is known to correspond to the diploid type, and a normal reduced germ cell of the blackberry.

5. The breeding data suggest that inR. vitifolius the female is the heterozygous sex.

6. The effects of introducing different proportions of parental chromosomes into hybrids are correlated with differences in morphological and other characters. This results in the intergradation of characters, a common feature of polyploids, and sometimes in a change of dominance.

In plums different pollinations can result in differences in the time of maturity and in the size of the fruit. The pollen influences the development of the fruit inasmuch as defective embryos promote earlier ripening and smaller fruits. Further, the wider the difference between the two parents, the more defective the embryos. The developmental differences in the fruit are an expression of different degrees of seed growth.

In some varieties of pears frost injury to the styles induces parthenocarpy, i.e. the complete absence of embryo development.

Difference in the fruit due to the action of the pollèn (formerly known as xenia) fall into three classes: (1) where the developing zygote is affected, (2) where the endosperm is affected, (3) where the effect is on the maternal tissues. In classes (1) and (2) the effects or differences are due to the action of paternal genes in heredity, and since they are readily explicable on a simple genetic basis there is no reason why they should continue to be referred to as xenia. In class (3) they are due to differences in the constitution and development of the embryo, endosperm and seed which affect the development of the maternal tissue. At one extreme the fruits are seedless (parthenocarpy), and at the other they have seeds larger than normal.