=34. Determinable and efficient differences.= The instruments employed in studying habitats can not be too exact, as there is no adequate knowledge as yet concerning the real differences which exist between related or contiguous formations. This is particularly true of differences which are efficient in producing a recognizable structural change in plant or formation. Investigations made by the writer have shewn that standard instruments will measure differences of quantity quite too small to produce a visible reaction. Efficient differences are not the same for different factors, and perhaps also for the same factor when found in various combinations. They vary widely for different species, being in direct relation to the plasticity of the latter. The point necessary to bear in mind in formulating methods for habitat investigation and in making use of instruments is that standard instruments should be used for the very reason that we do not yet know the relation between determinable and efficient differences. On the other hand, it is unnecessary to insist upon absolute exactness as soon as it is found that the determinable difference lies well within the efficient one. This by no means indicates that instruments are not to be carefully standardized and frequently checked, or that accurate readings should not be made. It means that a slight margin of error may be permitted in a machine which registers well within the efficient difference for that factor, and that instruments that read to the last degree of nicety are not absolutely necessary. In the fundamental work of determining efficient differences, however, instruments can not have too great precision. Moreover, these differences must be based upon the most plastic species of a formation, and the readings must be made under normal conditions. INSTRUMENTATION =35. Methods.= In the field use of instruments two methods have been developed. The first in point of time was the method of simple instruments, devised especially for class work, and capable of being used only where a number of trained students are available. The method of automatic instruments was an immediate outgrowth of this, due to the necessity which confronts the solitary investigator of being in different habitats at the same time. In the gradual evolution of this subject, it has become possible to combine the two methods in such a way as to retain all the advantages of the automatic method, and most of those of the method of simple instruments.