114 ARRANGEMENT AND STRUCTURE OF SUSTENTACULAR CELLS 



and a more superficial portion, the phalanx process, the former running from the 

 basilar membrane to the nucleated part of the hair-cell, the latter inclosed within 

 the region of, and between the acoustic elements. Gottstein (1870), Waldeyer 

 (1872), Nuel (1872), and Lavdowsky (1876) describe a true fusion of the apex of 

 the nucleated part of the supporting element with the lower portion of the corre- 

 sponding hair-cell, so as to produce a "twin or double cell." This view is to be 

 regarded as a misinterpretation, and from the investigations of Tafani (1882, 1884), 

 Retzius (1900), Held (1902), N. Van der Stricht (1908), and Kolmer (1910), it is 

 plain that in adult animals a close connection exists between these two elements, the 

 lower pole of the hair-cell occupying a cup-shaped depression in the upper pole of the 

 supporting cell body. The wall of this pit is made up of a system of deeply staining 

 fibrils which form a goblet- or chalice-shaped covering (Held) to sustain the sensory 

 epithelial cell. This chalice extends down into a broad, fibrillar filament running 

 obliquely toward the inner side of the nucleus; and coursing through the medial por- 

 tion of the cytoplasm to reach the base of the cell, where the filament enlarges into a 

 conical, fibrillated foot, the base of which rests upon the basilar membrane. At the 

 level of the nucleus the filament divides into two branches, one of which is connected 

 with the chalice, the other more slender one extending throughout the conical 

 phalanx process of the cell of Deiters, to abut against the membrana reticularis. 

 Parts of this filament have been noted by many previous observers. Its basal por- 

 tion has been erroneously interpreted by Boettcher (1869, 1872), Lavdowsky (1876), 

 and Nuel (1878) as a deep process given off from the neighboring hair-cell. The 

 chalice and the branch of the filament beneath it were not observed by Retzius 

 (1884, 1900), nor by v. Spee (1901), but these authors nevertheless describe a single 

 fibrillar band extending throughout the cell body and its phalanx process. The 

 chalice itself was first noted by Kishi (1902, p. 177), as "an dem unteren Ende der 

 Haarzellen befindliches Gebilde von Kelchformiger Gestallt, " but misinterpreted 

 as a nerve ending. 



The real seat of the nucleated body of the cell of Deiters is illustrated and 

 described by all authors in vertical radial sections, during the earliest stages in the 

 development of the organ of Corti, as alternating with that of the more superficial 

 hair-cells. Thus the supporting cell of the first row lies below an interstice sepa- 

 rating the acoustic elements of the first row from those of the second ; the supporting 

 cell of the second row lies below an interstice separating the acoustic elements of the 

 second row from those of the third; and the supporting cell of the third row lies 

 outside the acoustic elements of that row. No alteration in the relation of the sup- 

 porting elements to the hair-cells is mentioned. Even when representing each cell 

 of Deiters beneath its respective acoustic element, Retzius (1884, plates xxn and 

 xxm), and many others, make no allusion to it in their descriptions. N. Van der 

 Stricht is the only writer who has given an accurate explanation of these mutual 

 relations, which can be seen in sections tangential and somewhat oblique to the 

 surface of the epithelium, the two kinds of cells being represented by long, nucleated 

 columns, the real position of which can not be misinterpreted. According to his 

 investigations three successive stages are distinguishable: 



