If on the other hand we turn to the primary divisions of the vegetable kingdom, we find that certain large and natural groups came to be most generally recognised and placed in the front rank in every scheme; such were the groups of the Thallophytes, Muscineae, Vascular Cryptogams, Gymnosperms, Dicotyledons and Monocotyledons. But the coordination of these great divisions of the whole vegetable kingdom was far from being rightly understood. It was usage rather than anything else, which gradually put them forward as primary types; in the systems themselves some received too great, others too little prominence, or other groups of doubtful character were admitted alongside of them. Bartling, for instance, whose system up to 1850 or even longer may rank as one of the most natural, adheres to De Candolle’s division of the vegetable kingdom into cellular and vascular plants, and rightly divides the former into two main groups, Thallophytes and Muscineae (Homonemeae and Heteronemeae), while he separates the latter into Vascular Cryptogams and Phanerogams; but the Phanerogams are divided into Monocotyledons and Dicotyledons, which again are distributed into four groups, one of these being characterised by the presence of a vitellus, that is, of an endosperm surrounded by a perisperm,—a thoroughly artificial division. The three other divisions are named apetalous, monopetalous, and polypetalous, but the Coniferae and Cycadeae are placed in the apetalous division. Less satisfactory is the primary division into Thallophytes and Cormophytes proposed by ENDLICHER[39], the latter separating into the divisions Acrobrya (Muscineae, Vascular Cryptogams, and Cycads), Amphibrya (Monocotyledons), and Acramphibrya (Dicotyledons and Conifers); the names of the three latter groups, the first of which is utterly unnatural, are founded on erroneous assumptions respecting growth in length and thickness, which Endlicher borrowed from Unger. While Endlicher’s great work has continued down to our own time to be indispensable to the botanist as a book of reference on account of the fulness of its descriptions of families and genera, the system projected by Brongniart in 1843 has acquired a sort of official authority in France. The whole vegetable kingdom is here distributed into two divisions, Cryptogams and Phanerogams, and the former are incorrectly characterised as asexual, the latter as having distinction of sex. The Phanerogams, divided into Monocotyledons and Dicotyledons, are distributed into groups in a manner that is not satisfactory; but the system has one merit, that it keeps the Gymnosperms together in one body; and if they are incorrectly classed with the Dicotyledons, it was still a sign of progress, that Robert Brown’s discovery of gymnospermy was to some extent practically recognised. The system devised by JOHN LINDLEY[40] attained to about the same importance in England as attached to those of Bartling