Heredity 3:103 (1949)
Reproductive Versatility in Rubus

M. B. CRANE John Innes Horticultural Institution, Bayfordbury, Hertford, Herts
and P. T. THOMAS Department of Agricultural Botany, University of Wales, Aberystwyth

THE Veitchberry 4x = 28 and the Mahdiberry, 3x = 21, were both raised from a cross between the hedgerow Blackberry Rubus rusticanus, 2x = 14, and tetraploid forms of the raspberry Rubus idaeus, 4x = 28. No diploid hybrid, the result of crossing diploid forms of these two species, appears to be known and attempts we have made to cross them have entirely failed. It therefore seems that seeds are formed only when an unreduced germ cell from one or both diploid parents takes part in fertilisation or when a polyploid form of R. idaeus is one of the parents. That unreduced germ cells occasionally occur in R. rusticanus is evident, not only from the Veitchberry, but also from the origin of the John Innes Berry, 4x = 28, which was derived from R. rusticanus, 2x 14, crossed R. thyrisiger, 4x = 28 (Crane and Darlington, 1927).

In 1935 a plant of the Mahdiberry growing at Merton sent up a very vigorous basal shoot which was found to be pentaploid 5x = 35. This shoot was removed, but although care was taken it was not possible to be certain that it was attached to the original Mahdi. It is thus derived either as somatic mutation or by segregation in a seedling. Plants derived vegetatively from this shoot are much more vigorous, produce larger fruits and when provision is made for crosspollination they fruit more freely (fig. 1). The undersides of the leaves of the pentaploid are not so densely hairy as those of the triploid, but apart from this and the increase in size, which is proportionate in all organs, they seem indistinguishable.

FIG. 1.—The triploid Mahdi RRB, pentaploid Mahdi RRRBB and the tetraploid Veitchberry RRBB.

Most cultivated varieties of raspberry are heterozygous. Thus the variety Lloyd George with red fruits, red prickles, and hairy growth, gives seedlings with yellow fruits, green prickles, and glabrous growth when selfed. Similarly, the red-fruited variety Superlative with hermaphrodite flowers, in addition to giving yellow fruits, also gives male and female forms by segregation of recessive genes (Crane and Lawrence, 1931). Other genes which suppress pollen development also occur in raspberries (Lewis, 1939).

The Veitchberry combines the characters of its parent species. For example its fully developed leaves are partly pinnate as in R. idaeus and partly palmate as in R. rusticanus. Upon selfing, its offspring vary only in minor respects such as the colour of the petals and prickles; there is no approach to either of the parental forms. Upon crossing with R. idaeus and other species, the behaviour of the Veitchberry is again typically that of a species, but exceptionally a few aberrant forms have appeared in the progeny. All this suggests the general suppression of segregation, as in Primula kewensis, through the pairing of like chromosomes.

There are two forms of the Veitchberry; they differ in pollen development, one produces pollen abundantly, the other very sparsely and gives a proportion of the first form on selfing. Whether this difference arose somatically or sexually is not clear, but it is no doubt due to the action of the pollen-inhibiting genes found in the raspberry parent.

From crossing the Veitchberry as female with two diploid forms of R. idaeus: (1) the variety Lloyd George and (2) a male seedling from the variety Superlative, we raised two families, one of 118, the other of 126 seedlings. In both families variation occurred in petal and prickle colour and in other minor respects. But, apart from this variation, the result of the heterozygosity of the raspberry parent, 237 out of the 244 seedlings (97.2 per cent.) were intermediate between the parents, and of a uniform type in habit of growth and other major characters.

They were all highly sterile.

Several of them were cytologically examined and found to be triploid, 3x = 21.

The others, four plants in one family and three in the other were strikingly different. They were all of reduced size. Three of these are shown in plate . BB was very similar to its original blackberry grandparent R. rusticanus. RR was typically R. idaeus. RB was intermediate between these two species; it was like the Veitchberry but with slenderer and less robust growth, see fig. 2. The other four plants were three blackberry and one intermediate. The raspberry-like seedling was fertile, the others infertile, some of the blackberries entirely so.

The three illustrated seedlings were found to be diploid 2x = 14 and the other four must have been the same.

To elucidate the origin of these exceptional plants and assess the chromosome relationship between blackberry and raspberry, one of us (P.T.T.) examined the plant RB, the Veitchberry, and other allied Rubi shown below.

Veitchberry x Lloyd George 2x = 14 RB
Veitchberry x Lloyd George 3x = 21 RRB
Auto-triploid Raspberry 3x = 21 RRR
Mahdiberry 3x = 21 RRB
Auto-tetraploid Hailshamberry 4x = 28 RRRR
Veitchberry 4x = 28, pollen sparse RRBB
Veitchberry selfed 4x = 28, pollen abundant RRBB

The Veitchberry used in these studies was the sparse pollen form. The anthers usually fail to develop and those that do develop more or less normally contain only a small proportion of germ tissue isolated islands of pollen mother-cells are found amongst the somatic tissue. Evidently some physiological disturbance in the young differentiating anther causes some of the potential mother-cells to remain as somatic cells. The degree and type of chromosome pairing did not materially differ from that of the form with normal anther and pollen development which was also examined.

Chromosome association at meiosis in these two plants and in the auto-tetraploid raspberry Hailshamberry, are compared in table 1.


  Veitchberry 4x = 28   Hailshamberry 4x = 28
Types of configurations No. of
Types of configurations No. of
1 0 12 0 5 6 1 0 1 1
0 1 12 1 4 5 1 2 1 1
0 0 14 0 2 4 1 4 1 1
1 0 11 2 7 4 0 6 0 1
0 0 12 4 4 3 0 8 0 1
0 0 11 6 6 3 0 6 4 1
0 0 10 8 2 2 1 8 1 2
0 0 9 10 1 2 0 9 2 1
... ... ... ... ... 1 1 8   1
Total 12 4 356 103 31 Total  1 6 59 16 10
Average 0.4 0.1 11.5 3.1 ... Average 3.1 0.6 5.9 1.6 ...

It seems that the Veitchberry differs cytologically from the autotetraploid raspberry in three respects: (1) the number of quadrivalents is lower, 0.4 per nucleus in the Veitchberry, compared with 3.1 per nucleus in the raspberry, (2) the number of univalents is higher and (3) the range in number of univalents is also much higher in the Veitchberry than in the tetraploid raspberry.

Little is known of the way in which raspberry and blackberry chromosomes have become differentiated from one another, because no direct diploid hybrid between the two species has been produced and available for study. The information we have, has been obtained somewhat indirectly from the study of polyploid hybrids. For example in the loganberry, R. loganobaccus, which we have shown to have two sets of raspberry and four sets of blackberry chromosomes, there is little if any pairing between the raspberry and blackberry chromosomes (Crane and Thomas, 1940). On the other hand in the Madhiberry the degree of chromosome association is only slightly less than in an autotriploid raspberry (table 2).

Frequencies of cells with different numbers of bivalents or trivalents at meiosis in diploid and triploid forms of Rubus

Plant Formula 0 1 2 3 4 5 6 7 Total Average
Veitchberry 2x RB (II) 3 2 7 13 16 10 3 54 45
Mahdiberry 3x RRB (III) 7 5 11 15 10 6 6 6 66 33
Raspberry 3x RRR (III) 2 5 11 19 13 10 60 50

It, therefore, appears that raspberry and blackberry chromosomes are sufficiently differentiated to compel self-pairing when there are two sets of each in the hybrid but, when there is no opportunity for differential pairing as in the diploid, and to a lesser extent in triploids, the two types of chromosomes can pair with very little restriction.

Chromosome pairing in the Veitchberry is not that expected of a raspberry-blackberry tetraploid hybrid. Quadrivalents should be less frequent than in the auto-tetraploid but the wide range in number of univalents is not characteristic of an allo-tetraploid. We, therefore, suggest that the failure of pairing is genotypically controlled.

The exceptional diploid plants RB which occurred in the family Veitchberry x Lloyd George Raspberry enabled us to test this conclusion (table 2). There chromosome association ranges from complete pairing to almost complete failure of pairing. This kind of behaviour is what has been observed in pure species where the failure of pairing is due to the presence of a single gene. This interpretation would agree with our earlier conclusion that chromosome pairing in the Veitchberry itself is subject to genic interference.

The seven anomalous diploids must all have originated from the embryo sac of the Veitchberry, the male taking no part in fertilisation. In the BB plants it seems that the chromosomes from the Veitchberry were exclusively blackberry, in RR exclusively raspberry and in RB presumably one set of each.

Regular bivalent formation with autosyndesis (self-pairing) would give embryo sacs with the chromosome constitution RB. Haploid parthogenesis with this type of chromosome behaviour would thus lead to the intermediate type of plant. Segregation to give rise to embryo sacs of the constitution RR or BB for all seven chromosomes would demand complete quadrivalent formation with an orientation at the first metaphase of meiosis, such as would occur only once in 128 times. But our analysis of chromosome behaviour shows that quadrivalent formation is infrequent in the Veitchberry. Alternatively we may visualise the origin of the BB and RR embryo sacs as being derived from a random separation of unpaired chromosomes where there has been a high proportion of univalents.

FIG. 2.—Diploid blackberries BB, diploid raspberry RR, diploid intermediate RB and sexual triploid RRB, derived from Veitchberry RRBB crossed raspberry RR.


1. The Veitchberry 4x=28 was derived from R. rusticanus 2x=14 and a tetraploid form of R. idaeus 4x=28. Upon selfing there is no approach to either parental forms, it breeds true to its intermediate character.

2. Upon crossing with R. idaeus the behaviour of the Veitchberry is again that of a species, but exceptional diploid forms have appeared in the progeny: four blackberry-like, two intermediate but slender and one raspberry-like.

3. The chromosomes of the blackberry and the raspberry are differentiated and normally pair among themselves when two sets of each are present in the hybrid.

4. With less opportunity for differential pairing as in the Mahdiberry 3x=21 the chromosomes of the two types can pair effectively with one another.

5. The allotetraploid Veitchberry behaves exceptionally in that there is considerable failure of pairing. This failure appears to be due to genotypic control since no structural differences were observed between the raspberry and blackberry chromosomes.

6. One of the intermediate diploid plants from the Veitchberry similarly behaved like an asynaptic diploid.

7. It is concluded that seven diploids in the progeny from Veitchberry 4x=28 x raspberry 2x=14 have originated parthenogenetically from the embryo sacs of the Veitchberry.

Darlington Appendix

See also: Gustafsson: Differential Polyploidy within the Blackberries (1939)

Hurst Bibliography