FLORIN: SYSTEMATICS OF THE GYMNOSPERMS 355 



tion exudate still characterizes cycads, Ginkgo, taxads, and chlamydospernis, 

 and occurs in all conifer fanulies except the araucarians. Tiie ])ermian inver- 

 sion of the ovule was combined with a reduction of the single large air sac into 

 two separate smaller sacs, placed in such a position that the grain was brought 

 upwards through the micropyle with the germinal zone directed as before to- 

 wards the nucellus. From this stage, Dojde recognized two lines of development 

 in the Pinaceae. The primitive flotation mechanism was suppressed in both, but 

 in different ways. In the Araucariaceae, the grains fall on the cone scales and 

 develop long tubes growing towards the ovule — an advanced type of mechanism, 

 derived from the direct ovular reception type characterizing paleozoic conifers. 

 A significant trait of modern systematics is the combination of cytology and 

 taxonomy into cytotaxonomy (Anderson, 1937; Sharp, 1943). It is mainly the 

 number, morphology, and behavior of the chromosomes that are of importance. 

 The gymnosperms are remarkable for the stability of their chromosome numbers 

 and morphology. Lists of such numbers have been published by Sax and Sax 

 (1933), Sax and Beal (1934), Darlington and Janaki Ammal (1945), Sugihara 

 (1947), etc. The basic chromosome numbers of the genera in each family are 

 as follows: Cycadaceae 8, 9, 11, 12, 13; Ginkgoaceae 12; Araucariaceae 13; Podo- 

 carpaceae 12, 13, 19, 20; Cephalotaxaceae 12; Pinaceae 11, 12, 13; Taxodiaceae 

 10, 11; Cupressaceae 11 (other numbers uncertain); Taxaceae 11, 12; Ephedra- 

 ceae 7; Welwitschiaceae 7. The dominating numbers in the Cycadaceae are 8 

 and 9, in the Ginkgoaceae and Cephalotaxaceae 12, and in the Araucariaceae 13, 

 while most conifer genera belong either to a 12 series (Pinaceae) or an 11 series 

 (Taxodiaceae and Cupressaceae). The chlamydosperms, on the other hand, have 

 7 as their basic number {Gnetum unknown). Deviations from the primary basic 

 numbers have been explained by Sax and Sax, and Flory (1936) as being due 

 to the loss of one or more chromosomes following segmental interchange and 

 polyploidy, and to fragmentation and duplication of chromosomes resulting in 

 an increase in chromosome number. Under natural conditions polyploidy, though 

 not of a high valence, occurs in Picea (Kiellander, 1950), Pseudolurix and Juni- 

 perus (Sax and Sax, 1933), Sequoia (Stebbins, 1948), Ephedra (Florin, 1932; 

 Mehra, 1947), and Wehvitschia (Fernandes, 1936), but after colchicine treat- 

 ment of germinating seeds tetraploidy has also been brought about in Sequoia- 

 dendron (Jensen and Levan, 1941). The problem of the origin of the polyploids 

 found in the conifers has recently been discussed by Andersson (1947) and Steb- 

 bins (1950). 



Chemical characteristics are of value in the classification of gymnosperms, 

 but little use has so far been made of them. A pioneer work was that of Baker 

 and Smith (1910) on Callitris. Much later, Gibbs (1945) referred to the com- 

 parative chemistry of the Cupressaceae as one of the topics illustrating the use 

 of chemistry in taxonomy. The distribution of diterpenes of the phyllocladene 

 and podocarprene groups in the Podocarpaceae and Araucariaceae (Ilolloway, 

 1938), as well as the biochemistry of turpentines in the pines (Mirov, 1948) have 

 been studied from the point of view of phylogenetic classification. Erdtman 

 (1952) emphasized that constituents excreted into the dead conifer heartwood 

 as metabolic end products should be of special taxonomic interest because of 

 their indifference to external influences. Terpenoid constituents characterize 

 the heartwood of the Cupressaceae in contradistinction to that of the Pinaceae. 



