222 Pyrosoma Spinosum, Herdman 



of only 2 millimetres in length, which had not penetrated the inner 

 wall of the test. 



The musculature of the oral opening consists of a well-developed 

 sphincter, outside which run several more or less completely circular 

 muscle fibres. The atrial opening is not closed by a sphincter, but by 

 a horseshoe-shaped band of muscle, thinning away at each end. 



The atrial muscle is placed much farther forward than is usual, 

 lying across the centre of the branchial sac. It consists on either side 

 of a hollow fusiform band, split externally for the greater part of its 

 length. It is much longer and more slender proportionately in the 

 young than in the old ascidiozooids. It lies in the thickness of the 

 body-wall. There are also two transverse muscle bands situated in 

 front of the branchial sac. The most anterior is situated between the 

 endostyle and the mouth opening, and consists of a broad transverse 

 ventral band of muscle, whose ends break up into fine branches in the 

 lateral body-wall. The other band is similar and similarly branched, 

 and lies slightly anterior to the dorsal ganglion. 



The luminous organs are present as a patch of large circular 

 refractive cells, lying on either side of the body across the ciliated 

 band. There is a similar but smaller patch of cells on either side of 

 the atrial opening, which probably has a similar function. 



The dorsal ganglion has been figured by Herdman. It gives rise, 

 with others, to two pair of easily-observed nerves, the posterior of 

 which run to the upper ends of the two atrial muscles, and the anterior 

 pair to the anal muscle, giving rise to branches on the way which 

 could not be traced. 



The number of stigmata in a large ascidiozooid average about 

 forty, and the number of transverse bars about thirty. The relative 

 position of the stigmata and bars with reference to the oral aperture 

 and the oesophagus is rather unusual, as a line joining the two last 

 cuts across both stigmata and bars diagonally, while in the other species 

 of Pyrosoma such a line would be approximately parallel to the bars 

 and at right angles to the stigmata. This apparent distortion is due 

 to the wide distance which intervenes between the first of the dorsal 



