280 C. O. WHITMAN. 



of crescent-shaped orange-coloured spots on the dorsal side 

 of the lanceolate head. The concave side of each looks forward 

 and outward. The posterior eyes are three to four times as large 

 as the anterior ones. A few days later the eye-pigment has 

 I become dark brown. These eyes are sack-like involutions of the 

 epidemis (fig. 93, eight days after exclusion). Some of the 

 cells of the inner walls become very large and glassy, and are 

 connected, according to Leydig, with nerve-filaments. 



a. The Entoderm. — Whence arises the entoderm ? Thus far 

 we have found only two germ-lamcllse — ectoderm and mesoderm, 

 rig. 93 represents a sagittal section of an embryo eight days 

 after exclusion, but in about the same condition as stage 55 

 (7 ds.). This section is constructed from two successive sec- 

 tions, on one of which appeared the pharyngeal atrium [oa) and 

 on the other the anal aperture. All the diverticula are cut in 

 a plane a little one side from the middle. Only tw^o of the four 

 diverticula of the posterior region of the stomach appear {i.d.) . 

 Fig. 95 is a part of fig. 93 more highly magnified. The cajcal 

 cavities are still filled with the deutoplasm, or " residual yolk^' 

 (Laukester). The septa are composed of mesoderm cells, the 

 nuclei of which appear as mere dots. These walls are lined by 

 a loose layer of oval-elliptical cells (circa "01 mm.). In pre- 

 parations treated with osmic acid and hsematoxylin these cells 

 are very clear, and the deeply-coloured nucleus is very distinct. 

 The cells lie partially in the periphery of the yolk, the large 

 yolk-spheres being sparsely scattered through the cell-area. The 

 cells lie at some distance from one another (dorsal portion of 

 fig. 95), or in contact, with their longest diameter for the most 

 part parallel to the walls of the septa. Under a low objective 

 they appear as a light border around -the central field of yellow 

 yolk. At a little earlier date these cells (ent.) are even more 

 loosely arranged and intermixed with the peripheral yolk spheres. 

 Earlier still they are not to be recognised at all. One or two 

 days later (fig. 94, 9 ds. after exclusion) they are smaller, more 

 numerous, and compactly arranged in a single layer, with their 

 longer axes perpendicular to the septa -walls (fig 9-1 = horizontal 

 section). A longitudinal perpendicular section like that of fig. 

 93 proves that the formation of the entoderm progresses more 

 rapidly in the anterior and posterior than in the median region. 

 The intestine is still closed, and reaches totlu- anal aperture. A 

 little later this blind end becomes perforated, and thus the 

 alimentary canal is complete. 



b. Origin of the Entoderm.— \Miat is the origin of these 

 entoderm-cells ? Do they arise de novo, or have they a genetic 

 relation with the nuclei of the three primary blastomeres^ a, h 



