FLORIN: SYSTEMATICS OF THE GYMNOSPERMS 363 



laminated reproductive branching systems of early \ascular plants. Goebel 

 (1932) maintained his earlier conception, and Lanfer (1933) supported his 

 view. On the basis of ontogenetic studies of female conifer cones Ilagerup (1933, 

 1934) arrived at the conclusion that those of the Taxodiaceae, Pinaceae, Podo- 

 carpaceae, Araucariaeeae, and Cupressaceae are in the nature of inflorescences. 

 Exceptions are the Taxaceae and certain Juniperoideae. A cone and a long shoot 

 of the pine are similarly constructed. In the cone a short secondary axis develops 

 axillary to each bract. It carries two prophylls, and above them a median leaf 

 on the posterior side, which is the "ovuliferous scale." The construction is simi- 

 lar in other families, but the number of leaves on the floral axis varies. The in- 

 tegument is considered a megasporophyll ; it develops a basal megasporangium 

 (nucellus) on its morphological upper side. In the uniovulate flowers of Arau- 

 caria and Dacrydium, Hagerup postulated the presence of sterile prophylls, 

 while in Cupressus there would l)e no sterile leaves at all, and consequently no 

 ovuliferous scale. Juniper us has lateral as well as terminal flowers, inflores- 

 cences as well as single flowers; the sporophyll (integument) may be terminal 

 on the main axis. In the taxads, too, the integument was believed to be a sporo- 

 phyll. Already Lanfer (1934) however, criticized his interpretation of the in- 

 tegument, and so did later other morphologists, w^hile Emberger (1944) ac- 

 cepted it. Satake (1934), Doak (1935), Wettstein (1935), and Chamberlain 

 (1935) were adherents of the brachyblast interpretation. 



The persistent diversity of opinions convinced Florin (1938-1945, 1951) 

 that this problem could hardly be definitely solved by investigations on living 

 conifers alone, and that the fossil material so far considered had not been suffi- 

 ciently old to reveal the primary organization of their female organs. He there- 

 fore took up a comparative morphological study of the cones of fossil and living 

 conifers. It was found that the paleozoic cordaites and conifers furnished the 

 main clue to the interpretation of the cones of mesozoic and more recent coni- 

 fers. Primarily, the fertile complex is a radially SATumetrical short shoot in the 

 axil of a bract, and has the character of a strobilus or flow^er. In the most ancient 

 conifers (Lebachiaceae) the floral axis carries several sterile scales and one to a 

 few megasporophylls, each with one terminal orthotropous, erect ovule. The 

 younger types of cones have arisen by transformation of this primitive organi- 

 zation. The axis of the flower became reduced. Its symmetry was changed very 

 early, the megasporophylls becoming confined to the posterior side of the axis, 

 and the whole short shoot flattened. Disregarding for the moment the Ernestio- 

 dendron type, the paleozoic and mesozoic conifer cones are characterized by a 

 gradual modification and differentiation of the axillary complex into a fertile 

 part facing the cone axis, and a sterile part — the "ovuliferous scale." The sterile 

 scales and the sporophylls on the axis concurrently changed from a spiral to a 

 decussate arrangement. Not only the sporophylls, but also the sterile scales, 

 were moreover confined to the posterior side, including the flanks, of the flower 

 axis, while the anterior sector became wholly suppressed. The numl)er of sterile 

 scales was reduced until finally only one was sometimes left. In the paleozoic 

 Ernestiodendron type, too, the primary seed-scale complex was radially sym- 

 metrical, containing numerous sterile scales and a few sporoi)hylls, but in con- 

 nection with the flattening the sporophylls were favored at the expense of the 

 sterile scales. The reduction of the flower thus took a somewhat different course 



