1878. ] On the Natural Succession of the Dicotyledons. 725 
day; only two serious attempts having yet been made to cut loose 
from it, one of which has failed to command an acceptance, not- 
withstanding the eminence of its author, while the other is still 
too new to permit an opinion as to its ultimate success. I refer 
to Dr. Lindley’s classification as elaborated in his Vegetable King- 
dom, and to the system proposed by Sachs in his Tert Book of 
Botany. 
In the first} or, as it is supposed, highest of the three “ Divisions,” 
the corolla consists of distinct pieces or petals not at all connected 
with each other, and this is called the Polypetalous division. In 
the second group these petals are united, at least at their base, the 
distinct divisions only appearing, if at ali, in the form of a “limb”’ 
of separate lobes at the summit of a tube of greater or less 
length. This group is called the Monopetalous division. The 
third group wants the petals and corolla entirely, the perianth 
consisting of a single envelope which is always assumed to be the 
outer or calyx Plants of this nature form the Apetalous division; 
they are also frequently denominated monochlamydous, those of 
the other two divisions being designated as dichlamydous. 
The defects of this classification have long been apparent, and 
although so persistently adhered to, it has always been a source 
of trouble to systematists, no two ever entirely agreeing as to its pre- 
cise limits. Even those who wholly ignore or reject the doctrine of 
descent seek to bring together as nearly as possible those plants 
which actually resemble each other most, since this is the funda- 
mental idea, and formerly the only idea, of a “ natural system.” 
Yet in adhering to the principle involved in the classification by 
divisions this was frequently impossible, the two principles being 
often in direct conflict. In every such case it is clear that the 
Maintenance of the divisions involved, is so far, an artifictal, as 
contradistinguished from a xatura/ system of classification. And 
indeed it is difficult to see how a system of classification based on- 
the corolla is more natural or less artificial than one based on the 
stamens. If it could be shown that this class of characters are 
1 The propriety of commencing text books of Botany or Natural History with the - 
highest or most perfectly organized families and descending the scale so as to end 
with the lowest or most imperfectly organized may well be questioned. Though 
i doubtless more convenient for beginners to study, it” “tends to antagonize and keep- 
out of view the truth of a progressive historic ul evelopment in living things which, 
to the majority of students, is of greater value than any technical acquaintance bal o a 
_ Specialized forms. 2 
