TUMBLING NOT PAINFUL. 131 



stimulus. Still more conclusive, however, is the fact, so well 

 known to all Birmingham trainers, that if Rollers are allowed 

 unlimited liberty their performance deteriorates, and may even 

 altogether cease ; which is precisely what might be expected if, 

 as we have supposed, the epileptic tendency has been developed 

 by semi-confinement or other exciting causes. Unchecked 

 liberty in such a case, it might be supposed, would tend to 

 restore the nervous system to a normal state, and it does so. 

 We can also understand how many varieties of " Tumblers " 

 have lost, for want of cultivation, all propensity to tumble. 



Mr. Kesteven, however, ascertained one more fact, which 

 may almost be deemed final, and must certainly be so if sup- 

 ported by other observations. He submitted the brain of a young 

 Baldhead Tumbler to microscopic examination, and the result 

 was the discovery that the coats of the blood-vessels were 

 thickened, indicating irregular and unnatural blood-pressure. 

 These appearances are about the only constant sign of epilepsy 

 in the human brain. 



We regard it, therefore, as beyond any reasonable doubt 

 that tumbling is a symptom of some form of epileptic disease ; 

 but it by no means follows that all tumbling is entirely in- 

 voluntary or a cause of suffering to the birds. The same 

 nervous sensibilities which give most pleasure are capable, in 

 different degree, of causing severe suffering ; and it is also well 

 known that actions which are ordinarily of the character of 

 disease, by frequent repetition become habits, and afford a 

 measure of enjoyment, besides coming to a certain extent 

 under control. Even the line between voluntary and involun- 

 tary is not definite ; for taking a case strikingly akin to the 

 present, it is notorious that hysterical fits are in hundreds of 

 cases entered upon in a more or less voluntary manner, though 

 when unchecked they pass beyond the control of the patient 

 and occasion suffering. It is therefore conceivable that even 

 convulsive action of an epileptic nature may assume a form not 



