NO. 1783. NORTH AMERICAN LERN^OPODIDJE—WILSON. 205 



Similarly this gland shows a division into right and left halves along 

 the center. There arises out of it, on the midline between the two 

 halves, a homogeneous body which is at fii'st shaped exactly like a 

 musln^oom (ms., fig. 34). The umbrella portion at the anterior end 

 is about four times the diameter of the stalk, and lies with its rounded 

 top in contact with the inside surface of the nauplius epithelium at 

 the frontal margin. The stalk extends backward along the median 

 line and then curves over dorsally and terminates in a small peg (p), 

 which lies above and just behind the umbrella enlargement. 



This peg is the point of attachment between the filament and the 

 gland, through which the secretion of the latter is conducted into 

 the former (fig. 13). Every part of the organ, umbrella, stalk, and 

 peg, is hollow and is filled with the secretion from the gland. This 

 secretion is a stiff, homogeneous, adhesive substance, which hardens 

 into the filament and is indistinguishable from it in color and trans- 

 parency. 



In his description of the larva of AcJitheres percarum, Claus says 

 that he regards this organ as the duct of the gland which furnishes 

 the secretion. The present author is obliged to take issue with such 

 an interpretation for the following reasons: 1. The distal end of the 

 filament is completed at the very first, and subsequent growth takes 

 place at the proximal end, the new portion being pushed out against 

 that already formed and gradually coiled up to economize space. 

 The growing point of the filament is thus its point of attachment to 

 the gland. Ducts are not formed in this manner; they begin at the 

 gland and grow away from it, the growing point being at the distal 

 end. 2. The walls of the filament are absolutely homogeneous and 

 transparent, even when first formed, and show none of the structure 

 ordmarily found in ducts. 3. The filament is manufactured directly 

 from the secretion itself; gland ducts are never the product of the 

 secretion of the glands. 4. It is the filament which becomes the 

 attachment organ, and not the secretion which it contains, as will be 

 explained later. A gland duct that was afterwards torn away from 

 the gland and made to serve as an attachment organ would be an 

 anatomical novelty. 



Wlien this filament is first formed the stalk is comparatively short 

 and straight (a, fig. 14). But as the secretion accumulates the stalk 

 increases in length. Since the two ends are fastened at the very 

 beginning the only way for an increase in length to take place is 

 by a twisting and subsequent coiling of the stalk, and this is what 

 occurs. 



At first the stalk twists back and forth (b, fig, 14), or into a cork- 

 screw shape, and then, extending back into the tissue behind the 

 gland, it begins to coil. Eventually when fully developed it consists 

 of three large coils, which surround the rudimentary eye, and the two 



P 



