CiiAP. I. STRUCTUEE OF THE LEAVES. 5 



outwards, with their pedicels of a purple colour. Those on the 

 extreme margin project in the same plane with the leaf, or more 

 commonly (see fig. 2) are considerably reflexed. A few tentacles 

 spring from the base of the footstalk or petiole, and these are 

 the longest of all, being sometimes nearly i of an inch in length. 

 On a leaf bearing altogether 252 tentacles, the short ones on 

 the disc, having green pedicels, were in number to the longer 

 submarginal and marginal tentacles, having purple pedicels, as 

 nine to sixteen. 



A tentacle consists of a thin, straight, hair-like pedicel, carry- 

 ing a gland on the summit. The pedicel is somewhat flattened, 

 and is formed of several rows of elongated cells, filled with purple 

 fluid or granular matter.* There is, however, a narrow zone 

 close beneath the glands of the longer tentacles, and a broader 

 zone near their bases, of a green tint. Sjjiral vessels, accom- 

 panied by simple vascular tissue, branch off from the vascular 

 bundles in the blade of the leaf, and run up all the tentacles 

 into the glands. 



Several eminent physiologists have discussed the homological 

 nature of these appendages or tentacles, that is, whether they 

 ought to be considered as hairs (trichomes) or prolongations of 

 the leaf. Nitschke has shown that they include all the elements 

 proper to the blade of a leaf; and the fact of their including 

 vascular tissue was formerly thought to prove that they were 

 prolongations of the leaf, but it is now known that vessels some- 

 times enter true hairs.f The power of movement which they 

 possess is a strong argument against their being viewed as hairs. 

 The conclusion which seems to me the most probable will be 

 given in Chap. XV., namely that they existed primordially as 

 glandular hairs, or mere epidermic formations, and that their 

 upper i^art should still be so considered ; but that their lower 



* According to Nitschke (' Bot. (or tentacles) are coloured like 



Zeitung,' IbGl, p. 224) the purple parts of a leaf which do not fulfil 



fluid results from the metamor- their proper office." 



pilosis of chlorophyll. Mr. Sorby f Dr. Nitschke has discussed 



examined the colouring matter this subject in 'Bot. Zeituug,' 



■with the spectroscope, and in- 1861, p. 241, &c. See also Dr. 



forms me that it consists of the Warming ('Sur la Difference eutre 



commonest species of erythro- les Trichomes,' &c., 1873;, who 



phyll, " which is often met with in gives references to various publi- 



leaves with low vitality, and in cations. See also Greenland and 



parts, like the petioles, which Trt'cul, * Annal. des Sc. nat. bot.' 



carry on leaf-functions in a very (4tli series), torn. iii. 1855, pp. 



imperfect manner. All that can 297 and 303. 

 be said, therefore, is that the hahs 



