T HE purpose of this investigation was to study and correlate seasonal circulating ragweed antigen, which was de,scribed in a previous communication,l with thermostable antihody and the relation of each of these to reagin titer, symptoms, pollen count, and treatment with pollen extract. During this study, blood samples were obtained from both treated and untreated ragweed-sensitive patients and from normal, untreated individuals. No set schedule of treatment in respect to dosage or interval between injections was adopted. The pat,ients st,ndied received either the perennial or annual type of treatment, and varied from those extremely sensitive to those relatively insensitive. Considerable care was ta!:en in the evalntion pf symptoms and, in spite of detailed records, it was realized that appraisal of symptoms is always subjective on the part of hoth the clinician and the patient. Tndividual variations in exposure to pollen modify the improvement produced by specific pollen therapy. However, an evaluation of sympt,oms was attempted by reviewing each week a daily memorandum kept by the patient. A symptom score system was then adopted whereby symptoms for each w-eek were classified as 0, 1, 2, 3, and 4 in ratio to severity, and these were added as an accumulative score for the season for each patient (Table II). Blood samples were drawn at four-week intervals from March until August 15, after which the interval was shortened to one week. The samples were always obtained during each clinic visit prior to injection with pollen extract. Whole ragweed ext,ract was used for therapeutic injections. Circulating antigen, thermostable antihody, and reagin were determined on each blood within twent,y-four hours after it was received. This entailed the setting up and titration of several hundred tubes each week. For this rcason, the number of patients studied was limited to 29 treated cases and 11 controls. METHODS