482 PRINCIPLES OF GENERAL PHYSIOLOGY 



any new acquirement in reflex or association must be due to the formation of new 

 connections between neurones already present. Memory thus implies the more 

 or less permanent establishment of these connections. The possibility of dis- 

 connection at a later period must clearly be admitted. 



It appears that, in the lower organisms, such as insects, a habit may be formed 

 by long training ; so that, for example, they may become able to find their way 

 to food by a complex path. But suppose that the arrangement is altered back to 

 the simple one for a time and then the complex one, to which the new adjustment 

 has been formed, is returned to. It is clear that the length of time the new 

 acquirement lasts can be tested ; and experiments have been made on the cockroach 

 which show that about half an hour is the length of the time which this animal 

 is able to remember what it has learned. 



In the higher animals, new associations are formed, so far as we know, only in 

 the cerebral cortex. The experiments of Burnett (1912), already referred to, 

 showed that decerebrate frogs were unable to form even the simplest associations. 



In the lower animals there appears to be less centralisation. Yerkes (1912) 

 found that an earthworm, which had been caused to form a habit of taking a 

 particular course, did not lose the "memory" when the cerebral ganglia were 

 removed. 



With regard to the gratuitous introduction of such expressions as judgment, 

 or decision by some sort of a controlling " mind," which it has been thought by 

 some to be necessary to introduce even into the interpretation of the phenomena 

 shown by some of the simplest nervous systems, the experiments of A. A. Moore 

 (1910) on the starfish are to the point. The central nervous system of this 

 organism is in the form of a ring, from which a nerve passes radially to each arm. 

 If a simple cut be made across this ring, no break is made in the actual possibility 

 of control of each arm by the centre. The fact that the arm next the cut does 

 not co ordinate with the others in the righting movement proves that direct 

 nervous connection across the place cut is necessary for "intelligent" co-operation. 

 Any one arm can initiate impulses which affect strongly only adjacent arms and 

 rapidly decrease as they travel from their point of origin. Yet this simple 

 mechanism is sufficient to account for the complicated righting movements of the 

 animal. 



The paper by Carveth Read (1911) may be referred to in connection with the 

 relations between instinct and intelligence. 



SPEECH 



No reference has been made as yet to the nervous mechanism of speech and 

 those other powers, such as reading and writing, which depend upon it. 



Early in the evolution of social communities we find means of some sort for 

 the purpose of communication of signals of danger and so on. But very little 

 is possible with inarticulate sounds and it is only when articulate speech, with a 

 great variety of words having definite meanings, commenced that mental evolution 

 made rapid strides. 



For the cerebral centres and connections involved, the reader must be referred 

 to the textbooks of Human Physiology and especially to the monograph by 

 Mott (1910). 



METHOD OF INVESTIGATIONS 



In general terms, these methods have been described incidentally in the 

 preceding pages. They may be broadly divided into those of stimulation and 

 those of destruction of localised areas. The various histological methods of 

 staining different kinds of tissue and of degenerated tracts are also of importance. 



For accurate stimulation or destruction of localised spots in the interior of 

 the brain, the "stereotaxic instrument" of R. H. Clarke (Horsley and Clarke, 

 1908, pp. 19-39) is of great value. This has been recently improved, but details 

 of the latest form of the instrument have not yet been published. 



