OCTOBEK 1, 1897.] 



KNOWLEDGE. 



243 



periods ; yet numerous fossil plants of the Eocene forma- 

 tion of Europe were formerly referred to about ten of the 

 existing Australian genera. 



Returning to Australia, we find that although there are 

 Proteacea spread all over the country, from north to south 

 and east to west, two-thirds of the sis hundred species 

 are concentrated in West Australia, and these chiefly in the 

 extreme south-west. Moreover, less tban a score of these 

 four hundred species extend beyond the limits of that 

 colony. 



Fia. 2. — -Open Seed-vessel of Sikea crassifolia. Slight'y 

 reduced from natural size. 



The almost endless variety in habit, foliage, and flowers 

 presented by these plants fully justifies the botanical 

 designation of Pruteaccise. In the course of these articles 

 I may have mentioned the fact that the same general 

 appearance or aspect, the same kind ot foliage, and very 

 similar inflorescence, are repeated in the most distantly 

 related families of flowering plants, so that in the absence 

 of flowers or fruit it is often impossible to say to what 

 family a given plant belongs without a microscopic 

 examination of its anatomy — and this branch of botany is 

 still in its infancy. Classification is almost entirely based 

 on the structure of the flower, the seed, and seed-vessel 

 or fruit, and gross mistakes have been made by botanists in 

 naming plants from flowerless specimens. Instances are 

 not wanting in the Proteacew, for I remember one which 

 was originally described as a fern. 



A very large proportion of the Proteacew are shrubby, 

 varying from diminutive trailing plants to large bushes. 



.f\ 



%. 



vessel of Hitei 

 crrzS'iifoHfi. 



like rhododendrons and laurels. A few attain the dimen- 

 sions of large timber trees, and a few are almost herbaceous 

 in character. Almost all have thick, hard leaves, which 

 vary in size and shape to a surprising extent, and the 

 same range and kind of variation is repeated in difl'erent 

 genera. Often they are quite thick and woody, and they 

 usually persist for many years. The leaves of one resemble 

 a heath, of another an oak, of another a myrtle, of 

 another a fern, of another furze, and of another a thistle. 

 In many species, of various genera, the leaves are cylin- 

 diical, without any distinction of stalk and blade, and 

 having a formidably sharp point. In some they are reduced 

 to mere spines ; in others they are more 

 or less elongated and pliable. Huhi'ii 

 loirti, for example, has slender, whip- 

 like leaves, often more than a yard long, 

 and not more than a ouarter of an inch 

 thick. 



Coming to the flowers, we meet with 

 equal diversity in size, shape, colouring, 

 3nd arrangement, though the structure is 

 fairly uniform. Bright colours are not 

 uncommon, but several other families 

 surpass the ProUacea in brilliancy and 

 conspicuousness. Crimson and yellow pig. 3.— One of 

 are the prevaihng colours, and green with ^■^^g \^g geeds 

 :i 8lio-ht infusion of yellow is not infrequent, from ths Seed- 

 Tbe disposition or arrangement of the 

 flower is very striking in the genus 

 Banhsiti, which is, perhaps, the most 

 prominent of all the Proteacem — at least in certain dis- 

 tricts, especially western, outside of the tropics. In 

 this genus the flowers are crowded as thickly as they can 

 stand in spherical, 

 ovoid, oblong, or 

 cylindrical cones; and 

 as the axes of these 

 cones do not elongate 

 nor increase percep- 

 tibly in diameter, it 

 follows that there is 

 only space for a small 

 number of the seed- 

 vessels to attain full 

 development. Indeed, 

 out of five hundred 

 to fifteen hundred 

 flowers in a cone, 

 comparativelyfew are 

 fertile. In some of 

 the species the in- 

 dividual seed-vessels 

 are relatively large, 

 and then only from 

 a dozen to a score are 

 developed. Fig. 1 re- 

 presents a small cone 

 ill which five seed- 

 vessels are visible. Each one resembles a bivalve shell 

 inserted horizontally ; and under certain meteorological 

 conditions the dense woody valves open and permit the 

 seeds to escape. In some of the larger species the 

 cones are from nine to twelve inches long and from three 

 to four in thickness. A peculiarity of these and miny other 

 Australian trees and shrubs is the presence of many 

 generations of seed-vessels at the same time. This 

 phenomenon may sometimes be observed in other coun- 

 tries, but in Australia it is common. 



It is hardly necessary to explain that the genus Banksia 



Fig. 4. — Seed-vessel or Fruit of Cfre- 

 rillea qihhoaa^ begiuning to open. About 

 four -fifths of the natural size. 



