208 
REVIEWS. 
becomes elevated. Although, it remains uncertain whether or not the 
scale be the equivalent of a carpel, it does not necessarily follow that 
Gymnosperms must be deprived of a carpellary homologue. Lor if, in 
the case of the Larch and its allies, the ovule arise from an axial 
organ, the integument of the ovule may still be the equivalent of the 
carpel or carpels, as Baillon and others hold. If it be not carpellary 
then carpels are wholly suppressed. Leaving the question of the 
homology of the ovule-bearing scale, which we think a consideration 
of many other Gymnospermous genera (especially of the groups Cu- 
pressinesB and Taxinese) must show, need not form a very important 
element in the enquiry, let us briefly turn to the integument of the 
nucleus. Baillon and Dickson describe this integument as originating 
in two distinct lateral lunate elevations, one on each side of the small 
area to be occupied by the future nucleus. It is not clear whether 
the nascent nucleus or these lateral organs appear first. Much im¬ 
portance is attached to priority in the order of their development by 
both Baillon and his associates, and by Caspary. Baillon asserts the 
lateral organs rise first, but his figure scarcely supports him. 
It is difficult to believe that in this case it will ever be possible 
to obtain evidence of any real value. Lor granting that in all 
or nearly all cases examined the carpels appear before the ovule in 
Angiosperms, an exception is quite conceivable to any rule in vege¬ 
table organogeny, and an ovule appearing before its carpel need not 
surprise us. Indeed, Dr. Hooker shows that the nucleus of the ovule 
of Welwitschia actually appears before the perianth.* A more 
important point is the circumstance of the ovular integument 
originating in two distinct processes and not as a continuous 
ring in Abies. This affords to Baillon his strongest argument 
in favour of the integument being carpellary, each lunate ridge being 
a nascent carpel. The ovule thus becomes enclosed in a dicarpellary 
pistil, which terminates above in two prolongations representing 
styles. Caspary, in what appears to us the most important passage 
in his essay on the ‘ Lem ale Flower of the Abietinese,’ satisfactorily 
shows, however, the weakness of this evidence in favour of the com¬ 
pound nature of the ovular integument; showing that in some other 
Coniferae this integument is irregularly lobed, and, indeed, in the 
Larch, originates after the nucleus as a uniform surrounding ring, 
precisely corresponding in development with the ovular coat of An¬ 
giosperms. We cannot, therefore, admit on any organogenic or tera- 
tological grounds which have been hitherto advanced, the carpellary 
character of the ovular integument,—especially, if (to quote Dr. 
Hooker’s words, p. 31), “we assume the ovular integument of 
Gymnosperms to be carpellary, we must admit, first, that it has 
neither the form, structure, nor functions of an Angiospermous car¬ 
pel ; secondly, that it has those of an Angiospermous ovular coat; 
and, thirdly, that while the carpel is a singularly varying organ in 
Conf. pi ix. figs. 1—9. 
