South African Clawed Frog. 43 



one case, the deflection appears 8*3 a after stimulation ; in the other case, 

 the onset of the deflection occurs 11"8 a after stimulation, and is followed by a 

 large second deflection 4 cr later, i.e., lo"8 o- after the stimulus. This time 

 — 15'8 a — agrees "with other crossed reflex times from the same and other 

 decerebrate frogs, but ll'S o- is exceptionally short for a crossed reflex, and 

 agrees rather with the homonymous reflex times. There have been a 

 number of other cases recorded where the commencement of the deflection 

 shows a double character, a second deflection following the first by about 

 4cr. Caution must be observed in the interpretation of such curves 

 recorded by means of electrodes on the skin, where one has not point 

 leads on individual muscles. It is easy to understand how complicated 

 curves can be obtained from a single muscle when an electrode is not applied 

 exactly to the point where the motor nerve enters the muscle. In the 

 cases mentioned, however, the form of the curve, taken together with the 

 unusually short delay, rather suggests that, for some reason, the impulse, in 

 passing from one side of the cord to the other, has here taken a short path 

 which is not usually open in the decerebrate frog, and that another impulse, 

 traversing a path longer by the addition of one synapse, has later caused a 

 second discharge from the motor neurones. It may be that the short path 

 used here in the decerebrate preparation, as a rare exception, may be the 

 same path as is usually followed in the spinal preparation, where, as we have 

 seen, the heteronymous delays are, with frogs in good condition, similar to 

 the homonymous. 



If we attempt to analyse the average delay of the heteronymous reflex in 

 the decerebrate frog, assuming the delay at a synapse to be the same as in 

 the spinal frog, the following may be suggested : — 



Average delay of heteronymous reflex (35 records) 18 "So- 



Conduction in nerve (16 cm. at 35"5 metres per sec, 21'6° C.) ... 4"5 



Muscle latency 2 '4 



Latency of sensory endings 0'5 



Delay at three synapses (3 " l<r each) ll'l 



18 -.jo- 

 lt is doubtful whether any allowance need be made for the latency of 

 sensory endings. With the strong squeeze it is probable that the afferent 

 nerve fibres may themselves be stimulated. 



Homonymous Besponses in the Decerelrute Fray. 



The times of the homonymous responses determined in the decerebrate 

 frog are arranged in order of magnitude in diagram (fig. 5). The average 



