518 



SCIENCE 



[Vol. LV, No. 1428 



fewer in number than are those concerned in 

 the formation of the two main daughter nuclei. 

 The nuclear bodies formed by their fusion lie 

 ultimately lateral to the spindle instead of ter- 

 minal as in the case of the larger nuclei, and 

 are of strikingly small size. There may be as 

 many as four of the small nuclei at the end 

 of the first division of the pollen mother-cells. 

 When the second division takes place a further 

 formation of normal large nuclei (aggregating 

 four in number), and of abnormal small nuclei 

 more numerous than are the large nuclei re- 

 sults. The large nuclei give rise usually to 

 normal pollen grains but some or all of the 

 grains resulting from them may abort. The 

 small nuclei derived from the late-dividing and 

 small groups o-f ehi-omosomes give rise appar- 

 ently always to al)ortive grains. A number 

 of publications from this laboratory^ have em- 

 phasized the importance of pollen sterility as 

 a reliable moi"phological criterion of previous 

 heterozygosis or genetical impurity. 



Special attention has been devoted to abort- 

 ive pollen as evidence of hybridism in the case 

 of the Onagraceae and Rosaceea, but it is like- 

 wise found in many other groups. It is in- 

 teresting to note that TackhoLm in Sweden- 

 and Blackburn and Harrison^ in England, have 



1 Jeffrey, E. C, Spore Conditions in Hybrids 

 and the Mutation Hypothesis of De Vries, Bot. 

 Gas., Vol. 53, No. 4, October, 1914; Some Funda- 

 mental Morphological Objections to the Mutation 

 Theory of De Vries, American Naturalist, 1915. 



Standish, Ii. M., What is Happening to the 

 Hawthorns? Journal of Heredity, Vol. 7, No. 6, 

 June, 1916. 



Hoar, C. S., Sterility as the Eesult of Hybrid- 

 ization and the Condition of Pollen in BuJjus, 

 Bot. Gas., Vol. 62, No. 5, November, 1916. 



Forsaith, C. C, Pollen Sterility in Eelation to 

 the Geographical Distribution of Some Onagracete, 

 Bot. Gas., Vol. 52, No. 6, December, 1916. 



Cole, E. D., Imperfections of Pollen and Muta- 

 bility in the Genus Rosa, Bot. Gas., Vol. 63, No. 

 2, February, 1917. 



Jeffrey, E. C, Evolution by Hybridization, 

 BrooTclyn Botanic Garden Memoirs, 1: 298-805, 

 June 6, 1918. 



- Tackholm, Guuuar, On the Cytology of the 

 Genus Rosa. A Preliminary Note, Sartryck %ir 

 Svensh BotanisTc Tidskrift, Bd. 14, 2-3, 1920. 



recently pointed out the coincidence of hybrid- 

 ism and polyspory in the genus Rosa. Our 

 investigations have made this condition «lear 

 for a considerable range of Dicotyledons and 

 Monocotyledons. Tackholm has asserted on the 

 basis of his extensive studies that all the roses 

 belonging to the Canina section of the genus 

 Rosa, in other words, the roses of Europe, of 

 western Asia, and of northern Africa, are 

 throughout hybrids probably thousands of 

 years old and reproducing by apparently nor- 

 mal seeds, which are nevertheless formed 

 "apomictieally" and without the intervention 

 of a sexual act. Obviously such seeds will 

 "come true" as universally as do grafts or 

 vegetative multiplications of any kind and for 

 the same reason because they represent only 

 subdivisions of the vegetative body. 



Polyspory appears as a consequence of our 

 investigations, which will be published in full 

 at a later stage, as a frequent although not 

 invariable result of hybridization of species 

 (that is, of species crosses), and constitutes 

 one more valuable structural or morphological 

 criterion of heterozygosis. It frequently ac- 

 companies polyploidy and the manifestations 

 of the so-called "lethal factor" in marked re- 

 productive sterility in either known or sus- 

 pected hybrids between species of the higher 

 plants. 



We have now the following morphological 

 criteria of genetical impurity or heterozygosis 

 in plants, namely, reproductive sterility (most 

 easily observed in the case of the microspores 

 or pollen), gigantism, variability (mutability), 

 polyploidy and polyspory. Not all of these 

 may occur in any given case, but the coin- 

 cidence of any considerable number of these 

 features should be regarded as supplying 

 strong evidence of previous crossing of more 

 or less incompatible species or varieties. 

 E. C. Jeffrey 



A. E. LONGLET 



C. W. T. Penland 

 Laboratories op Plant Morphology, 

 Harvard IjNrvEESrrY 



3 Blackburn, K. B., and Harrison, J. W. H., 

 The Status of the British Eose Forms as deter- 

 mined by their Cytological Behaviour, An. Bot., 

 Vol. 35, No. 138, April, 1921. 



