PROCEEDINGS OF SECTION D. 323 



stem and growth of the twiner it^self. A close spiral may be produced 



by the combination of several Cassylha branches twining aide by side, in 



which case I have never seen any overlap, or production of haustoria 



laterally after contact and pressure due to mutual groTsi-h. It is a 



common occurrence with Cassytha to twine in horizontal and in any 



other direction. When the host axis is exhausted, the parasite ma}' 



grow approximately straight for lengths of over a metre, producing 



branches meanwhile. Finally, these horizontal branches, geotropically 



affected, droop to the ground from a low host, or remain pendant from 



taller shrubs, and may, after sprawling on the earth or hanging for a 



time, wither, or double back to the host. In some cases the stem or 



branch axis of the host is not accessible, and then, as in the case of 



Ulex europaea, there is produced a rude spiral of irregular diameter and 



direction as the haustoria penetrate — now here, now there — a phylloclade 



or exposed patch of stem ; but the general direction of the oblique 



spiral is reverted to after small deviations. When the twiiier has 



effected a normal spiral attachment, it loses sensitiveness to contact 



of the host for a time, but after a growth of at most 20 cm., the response 



to contact becomes so great that the end, circumnutating or not, readily 



twines round any object of small diameter up to about '8 cm., whether 



it be live or dead, or even metal, such as fencing wire. Frequently, 



when a Cassytha branch has outgrown the host twig and failed to find 



another or a new host plant by circumnutation, it turns back upon itself. 



and although, as previously stated, it was not affected by lateral or other 



contact with itself v/hile the host was available, the shoot responds now 



readily enough to stimulus afforded by contact with itself at a point 



that may be only a few centimetres farther back. There it at oncf 



twines, produces haustoria, and absorbs its own nutriment. That the 



haustoria thus produced create a physiological connexion was made 



evident in both laboratory and field. Vertical sections of haustoria 



showed that the cortical tissue of the older part had been penetrated 



by the conducting tissue of the haustoria which had made contact with 



similar tissue of the older portion of the stem. (2) Branches were cut 



above and below the spiral, and the lower end (of the axis only) , immersed 



in water coloured with eosin red, which stain in due course appeared 



at the end of the spiral, necessarily by way of the haustoria. (3) In the 



field, by cutting through a terminal loop, the returned branch end was 



rurjied into a separate graft on the living axis. In many cases the 



graft died owing to too great transpiration, coupled with insufficient 



penetration of the suckers, and perhaps to shock. Other cases lingered 



for days and slowly wilted, but a few survived, and continued to grow. 



For this retrograde movement, by which at first sight there nppears 



to be set up something kindred to the " vicious circle " known to 



zoopathologists, there is good and sufficient reason. With the twig or axis 



of the host outgrown and no fresh food material available, the nutating 



