The advantage of this plan appears to be, that the genera in each order, all depending upon modifications of the same parts, are of nearly an equal value, and the sections are only employed where there is a marked difference in subordinate characters or habit between the parts of a genus.

Some genera will therefore contain many of these sections, and others only one; in the same manner that an order is sometimes found to contain only one genus.

Or, perhaps, it would be more correct to say, that we are not yet acquainted with the other sections of the genus or the other genera of the order.

In illustration of this, I give the characters of the genera and sections (subgenera) in the tribe Thlaspideae, order Cruciferse.

According to the plan proposed by Mr Newman, the above sections must, I presume, be formed into genera, and the genera themselves will become families ; therefore the two (TeesdahVx and Platyspermum), which have only one section, must have new names invented, to mark the difference between the family and the genus.