8 GULLIVER, ON RAPHIDES. 
To show this, it will be sufficient to confine our attention, 
on the present occasion, to the leaves, and parts which are 
modifications of leaves, where we shall surely find raphides 
abundantly in certain plants, and at all seasons and in every 
variety of soil; while in other plants raphides ‘are as surely 
not so produced. Nay, even of two species belonging to 
different but closely allied orders, and growing close together 
in the very same spot of earth, the one will as constantly 
abound in raphides as the other will be destitute of them. 
But the last species may abound in spheraphides instead ; 
and these are so widely different from raphides that it would 
be a needless waste of time to repeat their diagnostics, 
which may be realised in a few minutes by any one who will 
compare them in such plants as Epilobium and Myriophyllum. 
Further, several species of one order, as Lilacee, may thus 
regularly differ in the presence or absence of raphides, of 
which Endymion and Allium afford curious examples. | 
Proceeding with our inquiry, it will be found that some 
raphidiferous orders may stand in the very centre of other 
orders not producing raphides. Thus, for example, as num- 
bered and expounded, from other characters, in Professor 
Balfour’s admirable ‘ Manual of Botany : . 
100. Loranthacez. 103. Valerianaceze. 
101. Caprifoliaceze. nek yen 104. Dipsacacez. 
Now, Rubiacez is not more distinguished here in print than 
in the type of nature as a raphis-bearing order, so far as my 
examination of the British plants has yet extended. In other 
words, I have never found a species of the central order 
without a plentiful crop of raphides, while search was vainly 
made for any traces of such crop in several species of the 
four surrounding orders. 
Further, with the instructive company of Professor Babing- 
ton’s ‘Manual of British Botany,’ we may view an order 
conversely ; that is to say, regularly devoid of raphides, and 
yet standing between two orders as regularly raphidiferous : 
80. Dioscorracex. 81. Hydrocharidacee. 82. OrcHipAcEgz. 
And here we shall find Hydrocharidacee differing as remark- 
ably in the want of those raphides in the possession of which, 
on the contrary, its next neighbours are so rich. 
Again, as for Monocotyledones, which are said in our books 
to abound so much in raphides, I have often examined all 
the grasses of a field, of which no less than eleven species 
were determined, besides four species of Carex, and never 
