THERMOMETERS 49 



Other Forms of Deep-Sea Thermometer. 



A thermometer which has done good service in the 

 past, and is still used by some observers, is the maximum 

 and minimum instrument (Fig. 7). It will not be 

 described here beyond saying that it depends on the 

 movements of two small indexes, which are pushed 

 along the capillary by the end of the thread of mercury. 

 Apart from the uncertain accuracy of the readings, 

 depending on the care with which the instrument 

 is made, it is impossible for it to give the temperature 

 when it does not change continually in one direction. 

 Suppose, for instance, that the temperature from the 

 surface to 50 fathoms is 15 or over ; from 50 to 100 

 fathoms, 10, and below that again it is 12. In this 

 lower layer the thermometer will fail ; the maximum 

 index will read the temperature of the upper layer, 

 15 5 , and the minimum index will stand at 10. Valuable 

 work was done with this instrument when no other 

 foim was available, but the limitations pointed out 

 make it useless for modern requirements. In the 

 instance given the temperature of the lower layer 

 could not be recorded unless the thermometer could 

 pass the upper layer without being affected by it, an 

 assumption which no observer is justified in making.* 



* Quite useful work may, however, be done with these 

 instruments at great depths in open water far from land, 

 where, so far as we know, no " sandwiching " of hot and 

 cold strata occurs except in the extreme south. To take 

 serial temperatures by their means, a fine stranded wire must 

 be used, not a rope of which the vibration jars the indexes 

 down. For a serial of, say, over 1,000 fathoms, take observa- 

 tions at 100, 300, 600 first ; then at 200, 400, 900 ; then at 

 500, 700, 800 and so on. After each observation plot the 

 temperatures on scale paper at proportionate depths ; an 

 observation which does not fall into the freehand curve drawn 



