most but not all of the fibers of these tracts terminated 

 before reaching; the thalamus. Walker (254) confirmed 

 in essence Wallenberg's observations and concluded 

 that the uncrossed fibers predominate and that many 

 terminate in the most medial portion of the arcuate 

 nucleus. Other workers (40, 117, igi, 192) reported 

 that the tract is uncrossed and that it terminates in 

 the arcuate nucleus or in the centrum medianum, or 

 in both. Recently Torvik (239) presented evidence on 

 the basis of retrograde cell degeneration that the 

 dorsomedial sector of the main sensory nucleus pro- 

 jects to the homolateral thalamus while the rest of this 

 nucleus projects to the contralateral side, thus con- 

 confirming some older observations (140) in this 

 respect. 



Since physiological evidence is conclusive that 

 both the contralateral and ipsilateral face areas are 

 represented in the arcuate sector of each ventrobasal 

 complex, it is tempting to assume that the uncrossed 

 tegmental trigeminal pathway exists and that it relays 

 tactile and kinesthetic impulses from the homolateral 

 face. Moreo\er, Hatschek's observation (117) that 

 the uncrossed tract is particularly prominent in un- 

 gulates would fit with the findings of Woolsey & 

 Fairman (277) that the ipsilateral cortical face areas 

 are unusually large in the pig and sheep. However, 

 there are also some reasons to doubt the correctness 

 of this assumption. First, it is obvious from the contro- 

 versy over whether this tract is both crossed and un- 

 crossed, only uncro.ssed, or whether it exists at all 

 (214), that different workers placed significantly 

 different lesions in their animals and there is no con- 

 vincing evidence that this tract necessarilv arises in 

 the main sensory nucleus. Second, it would be diffi- 

 cult to understand why homolateral tactile and 

 kinesthetic impulses should utilize a tract which is 

 quite different in its fiber composition from the ven- 

 tral pathway and why the homolateral tract should 

 lie so far apart from the contralateral one. Finally, the 

 usual observation of recent workers that the tract 

 ends in the most medial sector of the arcuate nucleus 

 does not immediately establish that it relays tactile 

 impulses, for in contrast to our own definition of the 

 arcuate component most workers include in it not 

 only the tactile thalamic trigeminal region but also 

 a ventromedial element (diff"ering considcraijly in 

 structure from the arcuate nucleus) which in our 

 opinion is not activated by tactile stimuli. It has been 

 suggested (207) that this element may be connected 

 with taste. It is of interest to point out that von 

 Economo (243) suggested a long time ago that the 



TOUCH AND KINESTHESIS 397 



dorsal trigeminal tract is in fact concerned with 

 gustatory impulses. 



Physiological Properties 



It has been known for a long time that destruction 

 of the posterior columns in man leads to a loss of the 

 capacity to appreciate the position and the movement 

 of the limbs, and to an inability to recognize the vi- 

 brations of a tuning fork applied over the bone. The 

 disturbances in tactile sensations were the subject of 

 some dispute. It seems reasonable to believe, how- 

 ever, that there is a severe impairment in the appre- 

 ciation of the spatial and temporal sequence of a 

 series of stimuli. In addition, increases in threshold 

 for tactile stimuli, a diminution in the number of 

 'sensory spots' and an impairment in proper localiza- 

 tion of the stimulus is often described. 



A general property of the lemniscal system is that 

 the information concerning the form, natiu'e, location 

 and temporal sequences of the impinging stimuli is 

 transmitted at each synaptic station with great secur- 

 ity. From the point of view of its organization, the 

 medial lemniscal system displays two striking features. 

 The first of these is that the peripheral sensory sheet 

 is projected centrally in a precise pattern, which is 

 preserved to a considerable degree through the suc- 

 cessive relays of the system, and is finally impressed 

 upon the postcentral cortex. The second is that the 

 system encompasses within a single tcjpographical 

 pattern several submodalities of the general sense 

 of mechanoreception. We wish to discuss the system 

 from these two \iewpoints. 



PROJECTION P.iiTTERNS IN MEDL-^L LEMNISCAL SYSTEM 



Patterns in Dorsal Columns 



The weight of the evidence indicates that the large 

 majority of ner\e fibers reaching the dorsal column 

 nuclei by way of the dorsal columns are axons of 

 first order neurons. It is not known that they are 

 exclusively so, however, and it is possible that some 

 unknown number arises from cells within the spinal 

 cord, cells which are activated by dorsal root afferents 

 and are therefore fibers of the second order. In the 

 cat some 25 per cent of the dorsal root myelinated 

 fibers which enter the dorsal columns at the seg- 

 mental level are believed to reach the cells of the 

 dorsal column nuclei directly (98). 



Examination of Marchi degenerations in the dorsal 



