INTRODUCTION. XV 



ing a piece of the dermal membrane, should be boiled 

 in nitric acid in a test tube, and carefully washed with 

 distilled water and the spicula mounted in Canada 

 balsam. Then the minutest of them will be rendered 

 accurately to the eye of the student, when sufficient 

 power is employed in the investigation, and in this 

 operation 600 or 700 linear is frequently necessary to 

 success. In the investigation of spicula thus mounted, 

 a caution is necessary, as there are frequently extraneous 

 spicula present which have either been incorporated in 

 the substance of the sponge, or attached to its surface, 

 and these are very liable to be mistaken for those of 

 the species under examination. It is therefore, a good 

 rule, never to consider a few such spicula present in 

 the slide under consideration as belonging to the sponge, 

 without the same forms can be detected in situ in the 

 slice of the sponge mounted in Canada balsam. Other 

 specific characters are used in the discrimination of 

 species, but those derived from the spicula, are by far 

 the most constant and reliable. It matters not 

 whether the sponge be young or old, perfect, or a mere 

 fragment, as long as these organs are present they 

 always exhibit their normal forms and proportions and 

 may be safely depended on in characterising either a 

 known species or in determining a new one, as in the 

 case of Desmacidon incognitus and some others. On 

 the contrary, form, size and colour of the sponge as 

 specific characters, excepting in a very few cases, are 

 perfectly deceptive, and although I have myself now 

 had nearly half a century's experience in the observa- 

 tion of British and exotic sponges, I frequently found 

 myself utterly unable to determine the species of a 

 mass of sponge put into my hands, but a slice from it 

 placed beneath the microscope frequently, at once, 

 solves the mystery. The student therefore, must not 

 attempt to jump to conclusions that the nature of the 

 subject renders morally impossible, but as a consola- 

 tion, I may venture to say that with the same degree 

 of care and steady investigation that he would exert in 



