228 THE SCOTTISH BOTANICAL REVIEW 



ring of sterile leaves may be united by symphysis, or may be 

 apetalous. The last condition is no indication of affinity, but is 

 simply a case of reduction. 



The pollen leaves (microsporophylls, stamens) lie between the 

 sterile and ovule leaves, and are largely adapted to the seed-producing 

 or pollen-distributing capacity. 



The ovules (macrosporophylls, carpels) occupy the highest portion 

 of the strobilus, being usually separate, and the carpels are often 

 separate as in Alisma, etc., forming apocarpia, or united as in lilies, 

 to form a syncarpium, with some portions still free as in the latter, 

 or with styles united all their length as in Commelinacese. 



When some other organs are attached to the syncarpia as in the 

 epigynous irids and orchids, "there has been such a fusion of the 

 originally separate parts of the strobilus as to result in a single 

 compact structure." 



" The primitive syncarpia of the monocotyledons appears to have 

 contained three carpels as in lilies." In the fusion some parts are 

 suppressed. In some dicotyledons, e.g. Caryophyllales, etc., " there 

 is a tendency to a reduction m the number of parts with increased 

 fusion of the strobilar leaves." 



" The same law is illustrated in the genetic line, which includes 

 the lilies (Coronarieae), pipeworts (Eriocaulaceae), sedges (Cyperaceae), 

 the lower grasses (Bambusese), and higher grasses (Agrostidae and 

 Panicese) ; or possibly still better in the line from lilies to amaryllids 

 (Amaryllidaceae), irises (Iridaceae), burmannias (Burmanniaceae), and 

 orchids (Orchidaceae)." 



There are then two modifications of the strobilus by {a) symphysis., 

 fusion of parts ; (b) aphanisis, a suppression of parts. 



They may affect some or all organs of the strobilus separately or 

 jointly. 



The conclusions arrived at from morphology are : — 



1. The identity of the cells and tissues of the two subclasses of 



Angiosperms indicates their close relationship. 



2. The fibro-vascular tissue-system of the dicotyledons indicates 



that the subclass is higher than the monocotyledons. 



3. The roots of dicotyledons indicate that this subclass is higher 



than monocotyledons, and suggest the possibility of the 

 origin of the former from the latter. 



4. The structure of the dicotyledonous stem indicates the highest 



rank of this subclass. 



5. Among monocotyledons the external morphology of the stem 



indicates the derivation from lily-like plants of the palms 

 and screw-pines by an excessive shortening of the internodes, 

 and of sedges and grasses by a corresponding elongation. 



6. The general morphology of the leaves of monocotyledons and 



dicotyledons, as has already been indicated many times, 

 emphasises the close relationship of the two subclasses, and 

 repeats the suggestion that the former includes plants which 

 must take rank below the dicotyledons. 



