CH. HI] INTERPRETATION OF APPEARANCES 91 



Appearances which seem perfectly unmistakable with a low power 

 may be found erroneous or very inadequate, for details of structure that 

 were indistinguishable with the low power may become perfectly evi- 

 dent with a higher power or a more perfect objective. Indeed the prob- 

 lems of microscopic structure appear to become ever more complex, for 

 difficulties overcome by improvements in the microscope simply give 

 place to new difficulties, which in some cases render the subject more 

 obscure than it appeared to be with the less perfect appliances. 



The need of the most careful observation and constant watchful- 

 ness lest the appearances may be deceptive are thus admirably stated 

 by Dallinger (see Carpenter-Dallinger, p. 427): "The correctness of 

 the conclusions which the microscopist will draw regarding the nature 

 of any object from the visual appearances which it presents to him 

 when examined in the various modes now specified will necessarily 

 depend in a great degree upon his previous experience in microscopic 

 observation and upon his knowledge of the class of bodies to which 

 the particular specimen may belong. Not only are observations of 

 any kind liable to certain fallacies arising out of the previous notions 

 which the observer may entertain in regard to the constitution of the 

 objects or the nature of the actions to which his attention is directed, 

 but even the mosf practiced observer is apt to take no note of such 

 phenomena as his mind is not prepared to appreciate. Errors and im- 

 perfections of this kind can only be corrected, it is obvious, by general 

 advance in scientific knowledge ; but the history of them affords a use- 

 ful warning against hasty conclusions drawn from a too cursor}' exam- 

 ination. If the history of almost any scientific investigation were 

 fully made known it would generally appear that the stability and 

 completeness of the conclusions finally arrived at had been only 

 attained after many modifications, or even entire alterations, of doctrine. 

 And it is therefore of such great importance as to be almost essential 

 to the correctness of our conclusions that they should not be finally 

 formed and announced until they have been tested in every conceivable 

 mode. It is due to science that it should be burdened with as few false 

 facts [artifacts] and false doctrines as possible. It is due to other 

 truth-seekers that they should not be misled, to the great w T aste of 

 their time and pains, by our errors. And it is due to ourselves that 

 we should not commit our reputation to the chance of impairment by 

 the premature formation and publication of conclusions which may be 

 at once reversed by other observers better informed than ourselves, or 

 mav be proved fallacious at some future time, perhaps even by our 



