The Development of Phascolosoma. ]^52 



above, though the stains and methods of staining- employed were 

 very numerous. Flemming's fluid for one minute, followed immedi- 

 ately by Orth's picrocarminate of lithium, and aceto-sublimate, 

 followed by Mayer's acid haemalum (or haemacalcium), are note- 

 worthy as producing slight, though still unsatisfactory, nuclear stains. 



The later stages of cleavage are more susceptible to differential 

 staining than the earlier, and fairly good whole preparations in 

 balsam can be made of the trochophores. 



For the study of cleavage I therefore recommend first of all 

 the living egg, in default of that, unstained eggs, mounted in 

 glycerine. Since the thick and highly-refractive yolk membrane 

 renders the microscopic image of the egg indistinct in balsam and 

 even in glycerine preparations, it may be removed by using a 3% 

 Labareaque's solution, which dissolves the membrane in about two 

 hours. If used with extreme care, with the eggs previously prop- 

 erly fixed, it will not injure the protoplasm, and may be found 

 useful. 



For staining sections of eggs, trochophores, and larvae I have 

 used principally Heidenhain's iron haematoxylin. For whole mounts 

 any good haematoxylin stain may be used, I have found a supple- 

 mentary staining with picric acid useful in differentiating the yolk, 

 with which the coelom of the larva is filled. 



For orienting and embedding embryos, I have found of great 

 value R. W. Hoffmann's (1898) modification of Patten's method, 

 which consists in orienting specimens on rectangular bits of glass 

 in droplets of a mixture composed of collodion and clove oil, 

 and fixing with xylol or benzole. 



