200 MR GEORGE BROOK ON THE 



while exceptionally small ones may only measure *85 mm. Boeck (2) gives the 

 diameter of the herring ova on the Norwegian coast as 1*5 mm. Those which 

 I have obtained from the Ballantrae herring averaged 117 mm. in diameter, 

 and those of the Loch Fyne herring are about the same size. The egg is 

 enclosed in an egg membrane, and outside the latter there is a viscous layer 

 by which the ovum adheres to anything with which it comes in contact. It 

 is by means of this substance that the eggs become attached to one another 

 in the form of flattened cakes. 



Viscous Layer. — If an isolated ovum is gently pressed from a ripe female and 

 examined immediately under the microscope, it will be found that the adhesive 

 material forms a comparatively even covering around the egg envelope. This 

 is even better shown if the egg be pressed out into dilute osmic acid solution, 

 or into spirit, and afterwards examined by means of sections. When thus 

 hardened the viscous layer usually appears structureless, or presents faint and 

 indistinct transverse streaks, such as would be shown in sections of a horny 

 substance. Fig. 1 shows a section of the viscous layer and egg membrane in 

 the unfertilised ovum. The relative thickness of the viscous layer varies very 

 much. When it forms a comparatively even layer around the egg membrane it 

 has usually about the same diameter as the membrane itself. When, however, 

 before hardening, the viscous substance comes in contact with that of another 

 egg, or with a foreign surface, a thickened welt is formed, which may be two 

 or three times the thickness of the egg membrane. In such cases there is a 

 corresponding thinning out of the layer in other parts of its circumference. 

 Although usually structureless, I have once or twice seen well-marked trans- 

 verse striations in the hardened viscous layer. Fig. 2 represents such an 

 appearance, and is taken from a ripe egg which was pressed directly from 

 the oviduct into osmic acid solution. The viscous layer has here about 

 the average thickness, but is divided longitudinally into two strata. Each pre- 

 sents a well-marked transverse striation, which is more distinct than that in 

 the inner portion of the egg membrane. The division of the viscous layer into 

 two strata may possibly be due to shrinking, but it appears difficult to see how 

 the transverse striations could be brought out by the same cause. They are 

 very evenly distributed throughout. 



This structure agrees with Hoffmann's (14) description of the appearance 

 of the viscous layer in the unripe egg. According to this author, the zona 

 radiata in the nearly ripe egg consists of two layers closely united together. 

 The outer layer, which corresponds to the viscous layer, is perforated by very 

 fine pore canals, and is sharply separated from the inner portion. In the ripe 

 egg, placed directly in l-10th per cent, osmic acid, the pore canals have almost 

 entirely disappeared. If the ripe eggs are, however, first brought in contact 

 with sea water, and afterwards fixed in osmic acid, the outer layer is seen to 



