6o 



HARDWICKE'S SCIENCE-GOSSIP. 



long, agile body and bushy tail, it bears a close re- 

 semblance to the weasel and stoat ; thus it is some- 

 times referred to the same genus. 



It also has the same pugnacious disposition as the 

 weasel, for we have a record from Delamere Forest 

 of a fierce encounter betwixt a female and a game- 

 keeper. It appears the man had taken its young to 

 destroy them, when the mother came too quickly on 

 the scene, and attacked the keeper. The fight con- 

 tinued nearly an hour : the polecat came off victorious, 

 for it escaped with its young, but the man was led 

 home blinded, and with his features lacerated in a 

 dreadful manner. Our hunt ended 

 far more happily, for we secured 

 the poor " fitchettj\ which has 

 furnished us with the text of the 

 present narrative. 



R. 



process of fertilisation ; but it was difficult to see 

 how this could be so, and moreover the great di- 

 versity of form in the irregularities seemed to nega- 

 tive such a supposition. On a closer examination, 

 however, and on sketching the stamens, I am inclined 

 to think that this is a case of abortion accompanied 

 by an extra formation in consequence of such abor- 

 tion ; that is to say, that the anther being in many 

 cases absent or imperfect, the energy of the stamen, 

 diverted from its usual object, has spent itself instead 

 in this unusual manner. To this conclusion I have 

 been led by observing that the extent of the malfor- 



ON THE STAMENS OF 



SPARMANNIA AFRICANA. 



O PARMA NNIA AFRI- 



w_) CANA belongs to the 

 Tiliaceae, or Lime-tree order, this 

 genus being a native of Southern 

 Africa. It grows there as a small 

 shrub, one foot or eighteen inches 

 high, with coarsely serrate, downy, 

 heart-shaped leaves, and umbels 

 of handsome white blossoms. 

 Each flower consists of an inferior 

 polysepalous calyx of four white 

 silky sepals, a hypogynous poly- 

 petalous corolla of four . white 

 petals, numerous hypogynous red 

 and yellow stamens, forming a 

 globular bunch, in the midst of 

 which is the style, rising from a 

 superior many-celled pistil. The 

 carpels are studded with tuber- 

 culated hairs, very much like the 

 glandular hairs of the stinging 

 nettle, but without the curved tip. 

 The pedicels exhibit a peculiarity 

 which I do not remember to have 

 seen elsewhere. About one-third of an inch from 

 each flower there is a joint, not very conspicuous, 

 but still easily seen, where the flower-stalk gives 

 way on being pulled. 



But the stamens form the most interesting part of 

 the plant. I have figured a few in order to convey 

 a better idea of their structure, by which it will be 

 seen that their filaments are more or less enlarged by 

 growths which sometimes take very fantastic shapes. 

 Puzzled at first to know what could be the use of 

 these formations, I not unnaturally expected to find 

 that they were in some way connected with the 



Fig- 58. — Flowers of S/ar»ianm'a Africana (natural size\ 



mations varies in the different stamens just in propor- 

 tion, roughly speaking, to the abortion of the anther ;. 

 and also that the outgrowths are largest at the end 

 of the stamen where the anther would have been, 

 and diminish in the other direction. That this is 

 the case will appear from the specimens figured, 

 which are fair examples of the rest. The antheriferous 

 stamens occupy by far the larger portion of the group, 

 being found in the centre around the style ; while the 

 abortive stamens are found towards the edge of the 

 group, forming a ring around the others. The latter 

 are comparatively short and are entirely yellow, while 



