macdougal: mechanism and conditions of growth 13 



auxo-thermal units, and that the enlargement is halting and ir- 

 regular, in some instances actual decrease of length and width 

 being apparent (see measurements in bold face type) ; and that such 

 shrinkage was not an error was amply attested by the auxano- 

 metric data to be given below. 



Many of the series of measurements of the growth of a single 

 bud extended over a hundred days. It was necessary to secure 

 rigid and secure mountings for plants which were to be kept under 

 observation so long. Preliminary arrangements for the experi- 

 ments to be carried out in 191 5 were made by making cuttings of 

 old joints of sound healthy plants. After the wounded surface 

 at the bases had healed, some of the joints were put in an equable 

 low temperature dark room until needed : others were set in pots 

 of sandy soil and others were set in dishes as water cultures. The 

 beginning stages of development were anticipated by cutting a 

 slot in a metal plate or disk through which the basal part of a 

 joint might be passed so that a small segment including the cal- 

 loused surface and roots, if formed, projected clear of the disk. 

 The joint of the cactus would now be securely fastened in place 

 in the disk by a setting of plaster. After this had become firm, 

 the preparation was firmly clamped to the top of an earthenware 

 pot or dish in such manner as to eliminate all possibility of any 

 vertical motion. The bud would be arranged to bear against 

 the lower side of a counterweighted auxanometer lever. This 

 instrument was likewise clamped to its base in such a manner as 

 to eliminate errors which might be caused by irregular movements. 

 The auxanometers consisted essentially of recording drums re- 

 volving daily or weekly, on which pens traced the movements of 

 improvised levers. These levers were arranged as simple or 

 compound, according to the needs of the experiment. The growth 

 of a joint from the initial 2 cm. to 12-15 cm. necessitated peri, 

 odical adjustment of the instrument. This was generally done 

 by blocking it up on its base and then securing by heavy steel 

 clamps (fig. i). 



Since some of the factors which would operate to cause changes 

 m external dimensions would also be operative in mature joints, 

 a number of these were mounted as above and put in contact with 

 auxanometric levers. Modification and control of the conditions 

 prevalent at Tucson were obtained by dark rooms, glass houses, 



