300 ME. H. M. BEENAED ON THE 



There is no other opening into the central canal. For a long time I was convinced 

 that a second pair of apertures existed in the 7th segment. A great accumulation of 

 faeces (PI. XXXIII. fig. l,f) is frequently found ovitside the central canal, far hack, near 

 the stercoral pocket, hut in adjacent diverticula. This, I concluded, must pass into the 

 hind-gut ^^itliout having to travel forward again all the way to the 1st segment. But 

 fresh series of sections have shown that there is no entrance here into the mid-gut. 



Behind this anterior pair of openings the tuhules which fill up the ahdomen open into 

 two lateral canals, which run backward more or less symmetrically from the anterior pair 

 of openings to nearly the end of the body (PI. XXXIII. fig. 1). The tubular diverticula 

 opening into these lateral canals show no regular arrangement, but branch out dorsally 

 and ventrally, /. e. above and below the great tracheal trunks which run along outside the 

 lateral canals (PI. XXXIII. fig. 5). They are arranged in irregular groups, alternating with 

 tufts of branches from the main tracheal trunks. In distended animals, the diverticula 

 hang down on each side of the great medio-ventral blood-sinus, which is continuous with 

 the neural aperture through the diaphragm (PL XXXIII. figs. 3, 4, n, and PI. XXXIV. 

 fig. 8, vs). This blood-space is especially large under the stercoral pocket in such cases, 

 and is here often found filled with a sohd mass of coagulum. 



The lateral canals run outside the series of dorso-ventral muscles. Posteriorly, they 

 are applied in a remarkable manner to the end of the mid-gut, being bound together 

 over it and under it by muscle-bands. The arrangement is seen in PI. XXXIII. figs. 1 

 & 7. "Where the lateral canal at this place is in contact with the mid-gut, its epithelium 

 is modified into a kind of supporting tissue (fig. 7, inep). It appears as if each cell had 

 become vacuolated, while its wall stiffened. Between this curious supporting epithelium, 

 and the actual wall of the mid-gut is a layer of fine blood-lacunae with coagulum, granules 

 (? digested matter), trachese, and {V) cells. It is in this part of the lateral canal that the 

 accumulations of faeces above mentioned are found (/), which led me at first to believe 

 that there must be an opening here into the mid-gut {mg). 



These lateral canals, therefore, supply a means of constricting the mid-gut just anterior 

 to its entrance into the hind-gut. The muscle-bands {tm) binding the two above 

 and below the mid-gut are connected by means of the rigid-looking, highly modified 

 epithelial cells of the inner walls of the lateral canals. The muscles with these epithelia 

 form a hoop round the end of the mid-gut. Some power of constricting the mid-gut 

 at this place is, as we shall sec, a necessary adaptation to the Arachnidan method of 

 feeding. 



The number and arrangement of the abdominal diverticula ia other Arachnids are of great interest, 

 inasmuch as they were originally segmentally arranged. 



Scorpio. — We find^ in serial sections as well as in dissection, five pairs of primary diverticula 

 leaving the central canal at regular intervals in the first five segments of the abdomen. As with the 

 first pair in Galeodes, these diverticula open into the mid-gut latero-dorsally. From their openings 

 into the mid-gut, the diverticula immediately branch out in all directions, forming the compact lobes of 

 the so-called liver. These lobes are strictly segmental, and are separated from each other by the dorso- 

 ventral muscles (PL XXXIV, fig. 1, dvm). The fifth pair of lobes run backward, filling up the rest of 

 the distensible portion of the abdomen. 



In this regular serial alternation of the diverticula with the dorso-ventral muscles, Scorpio has 



