IN NATURAL CLASSIFICATION. 601 
Lythracee. Rhamnee. 
Rosacee. Saxifragee. 
Leguminose. 
On the Hatr-Couuecrors op CAMPANULA, and the mode 
of its fecundation; by W. WıLson, Esa. 
(Tas. XX. B). 
Bv way of apology for the following remarks, opposed as 
they are in many respects to those of M. Adolphe Brongniart, 
in the “ Annales des Sciences Naturelles*,” it may be proper 
to state that this investigation was commenced with no- other 
object than to enjoy the gratification of witnessing the facts 
related by Brongniart, though with very little expectation of 
being able to follow him in all the details, much less of find- 
ing that anything bad been unaccomplished by him. I may, 
therefore, claim to have examined the subject with an unpre- 
judiced mind; and as I am fully sensible that authority of 
such weight ought to be treated with great deference, my in- 
quiries have, in consequence, been most carefully and delibe- 
rately conducted. The results are recorded with the more 
confidence, because they confirm, to a certain extent, the 
previous observations of Brongniart; although they lead to 
an opposite conclusion, and go very far to prove the accu- 
racy of the views held by Treviranus and Link. 
The principal fact insisted upon by Brongniart, is the re- 
tractile property of the hair (or pollen) collectors, immedi- 
ately after the period of fecundation. To the truth of this 
fact I can bear witness. The explanation offered by him, 
however, has no reference to the purpose for which so sin- 
gular a property exists. "The new facts which I. have ob- 
Served appear to throw much light on this point. 
The pollen-collectors in Campanula rotundifolia (the only 
Species which I have as yet examined), are thickly disposed 
kg Vol. 2; D. 244. 5 RE 
2x2 
