140 BULLETIN: MUSEUM OF COxMPAKATIVE ZOOLOGY. 



in the morning is more alkaline and contains a greater percentage of 

 solids than that collected at night. He considers that this is related to 

 the activity of the nervous system, and that it confirms Obersteiner's 

 theory of sleep. Halliburton (:01, p. 16) confirms Cavazzani's observa- 

 tions, and finds that in a state of fatigue there is a reduction in the 

 amount of organic solids in the cerebro-spinal fluid. 



The composition of the fluid varies, too, in diseased and normal 

 brains. The analyses of Halliburton (:0l) show that in spina bifida 

 there is a slight increase in the amount of iuorganic salts ; in hydro- 

 cephalus there is a considerable increase in the amount of pro- 

 teids. Mott and Halliburton ('98', :0l) have shown that, " in the 

 disease called General Paralysis of the Insane, the degenerative changes 

 that occur in the central nervous system are associated with the nres- 

 ence of the products of such degeneration in the cerebro-spinal fluid " 

 (:01, p. 437). Of these, choline and neurine, derivatives from the 

 breakdown of lecithin, a constituent of protagon, have a toxic action, 

 and when injected into the blood of another animal produce profound 

 functional disturbances. 



The cerebro-spinal fluid, Halliburton (:01) points out, has a double 

 origin, and it is questionable if its composition is the same in all parts. 

 He says : " It is found in the lymph channels and spaces of the brain 

 and cord tissue, and the perivascular lympliatics have been shown to 

 open into the sub-arachnoid space. It is found in the cerebro-spinal 

 cavity, and it can hardly be doubted that it is here formed largely by 

 the secretory epithelial cells which cover the choroid plexus." 



In all the analyses of cerebro-spinal fluid, so far made, the fluid has 

 been drawn from the meningeal spaces. These, being in connection 

 with the intraventricular spaces, it has been assumed that the fluid was 

 the same throughout. Fi'om the considerable number of characteristic 

 secreting areas in the walls of the brain ventricles, each wholly diff'erent 

 in structure, it seems probable that the cerebro-spinal fluid is a mixture 

 or connningling of at least several secretions. 



The researches of de Cyon ('98) and Johnston (:0l) have shown not 

 only that the saccus vasculosus is an organ for the supply of the cerebro- 

 spinal fluid, as Eabl-Rllckhard ('83) and others had previously sug- 

 gested, but that its epithelial lining contains ciliate sense cells which are 

 connected by nerve fibres with reflex centres, the whole constituting an 

 appai'atus Avhich probably controls the supply of the cerebro-spinal 

 fluid, and indirectly the blood-pressure and the heart-beat. 



In connection with this complex question of the function of the 



