METHODS IN BOLL-WEEVIL INVESTIGATIONS. 115 



ments have been placed about 35,000 weevils. Each compartment 

 presents varying conditions, with proper checks for testing the most 

 favorable and the most unfavorable conditions for the successful 

 hibernation of the weevil. 



At Dallas, Tex., a cage 20 feet square by 4 feet high (PI. II, fig. 4), 

 having 4 sections, has been built at a cost of slightly over $20, for 

 additional hibernation tests and also to serve as a check upon the 

 ex2:)eriments at Keatchie, La. In the Dallas cage have been placed 

 more than 10,000 weevils, and as many more have been distributed 

 in smaller cages at various points in Texas. From the large number 

 of conditions represented and the more than 50,000 weevils under 

 test, it is hoped that valuable data may be obtained upon the hiber- 

 nation of the boll weevil. 



In the Avork done in the field cages it was found that whenever pos- 

 sible each individual in a lot should be so marked as to be easily 

 recognizable. The definiteness of the record work was also largely 

 increased by the practice of attaching a tag to each square or boll as 

 it was attacked by a weevil. Upon the tag was recorded all data 

 as to date and nature of attack. Daily examinations were then made 

 of all these tagged squares or bolls and additional data recorded as 

 it developed. In this Avay the full data for thousands of cases have 

 been obtained and from it many important conclusions have been 

 drawn. The tags in any series of records are numbered consecutively, 

 that it may be possible to know when the series is complete and to 

 give at any time an approximate idea of the number of records 

 already obtained. If a number of experiments with varying condi- 

 tions are to be made, each portion of the experiment is described and 

 given a serial number by which it may be designated upon the tags. 

 After the experiment is completed the data from the tags is trans- 

 ferred to permanent record cards for incorporation in the card cata- 

 logue system. \\Tierever possible the data is recorded in tabular 

 form and the tables followed by cards giving a full summary of the 

 results obtained. 



The record cards for general note work are 4 by 6 inches in size. 

 The sheets are arranged in books of 50 leaves each, making the con- 

 venient pocket size of 4 by 7 inches. Perforations across the bases of 

 the sheets make it easy to remove them as desired. Across the left- 

 hand end of the sheet is space for the name of the Avriter of the notes. 

 At the top of the sheet are spaces for " Su|>ject,"' '' Date," and 

 " Locality." The body of the page is cross ruled, giving 12 by 22 

 lines for the records. These cards have been found adaptable for 

 practically all the permanent record work of the investigation. The 

 cards have been arranged by subject and alphabetically, but it is our 

 purpose to arrange the entire catalogue upon a decimal system during 

 the present winter. 



