This paper presents a study on the generation of mathematics test items using algorithmic methods. The history of this approach is briefly reviewed and is followed by a survey of the research to date on the statistical parallelism of algorithmically generated mathematics items. Results are presented for 8 parallel test forms generated using 16 algorithms covering a variety of mathematics content and cognitive categories. The majority of the algorithms yielded items that were very homogeneous in their statistical characteristics. Those algorithms that did not yield homogeneous items were analyzed to determine if causes for differences could be determined. A possible innovative application of the algorithms is the computer generation of new test forms with specific content and statistical specifications and without the need for a preexisting item bank. Includes two tables, three figures. (Contains 16 references.) (Author)