INHIBITION AND CARDIAC VAGUS 109 



according to Stefani is increased by vagal excitations, 

 must obviously in some sense be unrelaxed. Because an 

 inactive voluntary muscle appears soft it by no means 

 follows that the lengthened cells are themselves so. They 

 have merely ceased to pull on their origin and insertion. 

 One fibre is not attached to another, and in contraction 

 what we observe is the general tensile strain. But in the 

 heart this condition of loose muscle fibre does not exist. 

 According to Schafer cardiac fibres differ greatly from 

 voluntary fibres : " their striations are less distinct ; they 

 have no sarcolemma ; they branch and unite with neigh- 

 bouring fibres, and their nuclei lie in the centre of the 

 fibres." Of these differences, and here I follow a brilliant 

 suggestion of Keith's, the really important one is that 

 " they branch and unite with neighbouring fibres." If the 

 diastolic muscle action really does work, the fibres cannot 

 slacken and bend as in voluntary muscle. They form an 

 actual network, an interdigitated or branched growth- 

 mechanism, and must move together. Excitation is not 

 transmitted from fibre to fibre in skeletal muscle, but it 

 does so pass in cardiac as in smooth muscle, and all the 

 cells are excited in waves. It is obvious that these millions 

 of short columnar cells, each with its restraining connections, 

 have to act as a body, cell with cell, fibre with fibre, and 

 layer with layer, since the layers are so much part of each 

 other that anatomists differ as to their number. In such a 

 formation we have a most remarkable and unique engine, 

 very different indeed from the fibres of voluntary muscle 

 isolated in their sarcolemma, and a rough illustration of 

 its mechanism may be afforded by comparing it with what 

 we see in the instrument known as a " lazy tongs," in which 

 interdependent, interbranched, and hinged lozenges are 

 shortened or lengthened at the user's will, while in either 



