for example, the badges of the Regiment of Light Artillery (1812-1821). These gaps 

 will be filled only by excavating at sites known to have been occupied by specific 

 Regular units during particular periods. Indeed, since this study was begun, four 

 unique and significant insignia were excavated at the site of a War of 1812 canton- 

 ment, and these greatly enrich our knowledge of the period. 



The record of insignia of the veritable multitude of independent uniformed 

 Militia companies in existence during the period under consideration may never be 

 complete. The selection presented here, however, is an excellent representative 

 chronological cross section of typical designs and variations of insignia worn by the 

 uniformed or "volunteer" Militia, as opposed to the "common" or "standing" 

 Militia. 



The best sources of documentation and dating for Regular Army devices are 

 the uniform regulations and ordnance regulations; these are supplemented by perti- 

 nent records in the National Archives, notably the letter files of the Purveyor of 

 Public Supplies and of the Commissary General of Purchases. The letter files are 

 voluminous, but in some cases badly mixed and in many cases incomplete. We have 

 conjectured a reason for this incompleteness. The two prime contractors for mili- 

 tary insignia during the period 1812-1821 were George Armitage and William 

 Crumpton, both of whom had their small factories in Philadelphia within a mile of 

 the office of Callendar Irvine, Commissary General of Purchases. The paucity of 

 written transactions in the records in the National Archives between these gentlemen 

 and Irvine tends to bear out our assumption that most of their dealings were con- 

 ducted verbally in Irvine's office. This would account for the lack of sketches and 

 drawings of cap plates and belt plates in files of the National Archives. In cases 

 where no specific documentary evidence is available, dating has been based on a 

 careful evaluation of design development and comparison with biographical speci- 

 mens that can be more fairly dated through knowledge of the former owner's career. 

 Excavated insignia from datable sites have also reduced the problem considerably. 



For Militia insignia worn about 1835, the best documentation is to be found in 

 U.S. Military Magazine, published between 1839 and 1842 by Huddy and Duval of 

 Philadelphia, and in New York Military Magazine, published by Labree and Stockton 

 of New York during 1841. In 1939, Frederick P. Todd described the Huddy and 

 Duval prints in detail {Journal of the American Military Institute, 1939, vol. 3, no. 3, 

 pp. 166-176). However, evaluation and consideration of over-all design develop- 

 ment and comparison with dated biographical specimens of the earlier period, before 

 1835, are difficult and must be done cautiously, as there is no orderly pattern. One 

 generalization does seem clear: during the decade after 1821, when the Regulars 

 discarded large cap plates, the Militia almost universally adopted them and con- 

 tinued to wear them well into the 1840's. Very few insignia include the maker's 

 name or initials, but when they do, bracketing within a definite period is relatively 

 easy. Similarly, when a cap plate appears to be original to a cap, the design of the 

 cap and its maker's label, if included, are of great help. Finally, when there is 

 nothing else to rely on, the "feel" of the specimen, gained through the experience 

 of studying several thousand, has been used, although with reluctance. 



The year 1800 was selected as the opening date of the study because it was in 

 that year that the first metal ornament was prescribed to designate a particular 

 branch of service. The closing date of 1851 was chosen because Regular Army 

 devices for that year and thereafter are well documented in uniform regulations, 

 manuals, and catalogs of manufacturers such as William Horstmann and Sons. 

 Militia dress after that general date becomes so increasingly complex that it should be 

 attempted only as a separate study. 



