OBJECTIVE: To investigate if and to what extent typical head pain can be reproduced in tension-type headache (TTH), migraine without aura sufferers, and controls when sustained pressure was applied to the lateral posterior arch of C1 and the articular pillar of C2, stressing the atlantooccipital and C2-3 segments respectively. BACKGROUND: Occipital and neck symptoms often accompany primary headache, suggesting involvement of cervical afferents in central pain processing mechanisms in these disorders. Referral of head pain from upper cervical structures is made possible by convergence of cervical and trigeminal nociceptive afferent information in the trigemino-cervical nucleus. Upper cervical segmental and C2-3 zygapophysial joint dysfunction is recognized as a potential source of noxious afferent information and is present in primary headache sufferers. Furthermore, referral of head pain has been demonstrated from symptomatic upper cervical segments and the C2-3 zygapophysial joints, suggesting that head pain referral may be a characteristic of cervical afferent involvement in headache. METHODS: Thirty-four headache sufferers and 14 controls were examined interictally. Headache patients were diagnosed according the criteria of the International Headache Society and comprised 20 migraine without aura (females n = 18; males n = 2; average age 35.3 years) and 14 TTH sufferers (females n = 11; males n = 3; average age 30.7 years). Two techniques were used specifically to stress the atlantooccipital segments (Technique 1 - C1) and C2-3 zygapophysial joints (Technique 2 - C2). Two techniques were also applied to the arm--the common extensor origin and the mid belly of the biceps brachii. Participants reported reproduction of head pain with "yes" or "no" and rated the intensity of head pain and local pressure of application on a scale of 0 -10, where 0 = no pain and 10 = intolerable pain. RESULTS: None of the subjects reported head pain during application of techniques on the arm. Head pain referral during the cervical examination was reported by 8 of 14 (57%) control participants, all TTH patients and all but 1 migraineur (P < .002). In each case, participants reported that the referred head pain was similar to the pain they usually experienced during TTH or migraine. The frequency of head pain referral was identical for Techniques 1 and 2. The intensity of referral did not differ between Technique 1 and Technique 2 or between groups. Tenderness ratings to thumb pressure were comparable between the Techniques 1 and 2 when pressure was applied to C1 and C2 respectively and across groups. Similarly, there were no significant differences for tenderness ratings to thumb pressure between Technique 1 and Technique 2 on the arm or between groups. While tenderness ratings to thumb pressure for Technique 2 were similar for both referral (n = 41) and non-referral (n = 7) groups, tenderness ratings for Technique 1 in the referral group were significantly greater when compared with the non-referral group (P = .01). CONCLUSIONS: Our data support the continuum concept of headache, one in which noxious cervical afferent information may well be significantly underestimated. The high incidence of reproduction of headache supports the evaluation of musculoskeletal features in patients presenting with migrainous and TTH symptoms. This, in turn, may have important implications for understanding the pathophysiology of headache and developing alternative treatment options.