346 MORPHOLOGY. 



terms stamina abortiva and stamina castrata have no meaning. 

 In that relation it corresponds completely to the sporophyll of the 

 cryptogamic stem plants, and the forms there exhibited as ty- 

 pical of classes are here again manifested to characterise families 

 or genera. 



We find here the sporophyll of most Ferns, developing a number 

 of capsules (here termed cells, or loculi), out of the under face of 

 the leaf, in the Cycadacea. In many Coniferous plants, only a few, 

 long and tubular loculi are formed on the under surface (as in 

 Cunninghamia). In Juniperus, Cupressus, &c., the stamens can- 

 not be distinguished at all from the sporophyll of the Equisetacece ; 

 and we find in Humirium and Glossarrhen, where, however, two 

 loculi are presented instead of one, an analogy with the sporophyll 

 of the Lycopodiacea, where a capsule is formed on the upper 

 surface of the base of a flat foliar organ. But the stamen 

 usually corresponds to the sporophyll of the other Ferns, in which 

 only the petiole and mid-nerve of the leaf are perfected, at the 

 sides of which the parenchyma merely forms the loculi. The 

 structure, however, corresponds not to the much divided Fern 

 leaf, but usually to a simple flat leaf, with a petiole. Then it 

 exhibits an attenuated basis (the petiole is here termed Jilamenf), 

 and an upper broader part, the blade of the leaf (here termed 

 antherd). In the anther, we further distinguish a middle part (the 

 mid-nerve of the leaf, here termed the connectwum) from the 

 lateral parts, the chambers (loculi or thecce) which appear at the 

 summit, the edges, the upper or under surface of the connective as 

 globular, oval, or long cylindrical projections ; besides these, the 

 original edge of the leaf as a longitudinal furrow (rima longitu- 

 dinalis]. Finally, in many stamens the entire leaf substance, in 

 analogy with the so-called sessile leaf, is applied to the formation 

 of pollen-chambers (anther a sessilis). 



Each stamen originates as a leaf, runs at first through a similar 

 series of forms, and its subsequent peculiar appearance is merely a 

 result of its special mode of development, which may be traced, 

 not merely ideally, but mostly really, in the progressive develop- 

 ment, to a few simple fundamental types. Besides the cryptogamic 

 type, above followed out in the families of the Cycadacece and Coni- 

 ferce, a Phanerogamic type is also to be traced ; which essentially 

 consists in the circumstance, that, independently of the presence of 

 the filament, a flat leaf is so developed that its mid-rib becomes the 

 connective, its edge the longitudinal furrow, its parenchyma swells 

 out on both sides of the connective, in which then, through the 

 formation of the finally free pollen-grains, one (as in Abies and the 

 Asclepiadacece) or commonly two thecae are commonly formed on 

 each side. This type doubtless lies at the base of all Phanerogamic 

 stamens, if we except Najas, Caulina, and some Aracea (of which 

 I do not know the history of development). All further peculiar- 

 ities concern either the development of the theca3 on one side alone 



