DISCUSSION

There was a significant decrease in oxygen consumption and breath rate after both cyclic meditation (a technique which included yoga postures interspersed with periods of supine rest) and after SH (supine rest alone). The breath volume was raised after both practices. The magnitude of change for all three parameters was greater after cyclic meditation and did not vary with the duration of experience of shavasan and of cyclic meditation, though there was a wide range in this experience among the subjects (i.e., from 3 to 72 months).

The breath rate. depth, and breath phase duration(inspiration/expiration)are all highly sensitive to phasic changes in the psychological state(Lorig & Schwartz. 1990). Theoretical models of pulmonary response have tended to identify a single variable, typically the metabolic rate, and to attribute the majority of changes in pulmonary response to that variable. For example, early studies on Transcendetal Meditation reported that the observed feduction in metabolic rate (and hence in the need of oxygen) during meditation was reflected in adecrease, essentialaly involuntary. in the rate of respiration and in the volume of air breathed (Wallac & Benson, 1972). The greater decreased in oxygen consumption, decrease in breath rate, and absence of change in minute ventilation on spite of the greater breath volume) after CM, may have a similar explanation.

The greater reduction in oxygen consumption after CM and the changes in respiration suggested that in spite of the practice of yoga postures, which may be expected to be more stimulating than supine rest is, the effect suggested relaxation. These findings were opposite to a previously reported increase in oxygen consumption following virasana, a standing yoga posture (Rai & Ram, 1993). This standing yoga posture induces a hypermetabolic state with increased sympathetic activity, which disappears when the subjects adopts a supine posture (SH). The present results also suggest that while practicing yoga postures for relaxation, two factors are important, maintaining a relaxed mental attitude and having periods of rest (no activity) following a period of activity.

The importance of interspersing exercise with periods of rest has already been described (Falk, 1995). Intermittent exercise was described as more likely to enhance enjoyment and improve compliance with the exercise plan. Yoga postures have been shown to serve as a form of mild exercise (Rai & Ram, 1993). The yoga postures practiced in CM may not have served as a form of exercise, through possivly of smaller magnitude , has not been evaluated. However, it as been shown that another mind-modyfying Oriental practice, Tai Chi, which has been described as a "moving meditation," caused changes in heart rate, blood pressure, and urinary catecholamines, which were similar to walking at speed of 6Km/ hour(Jin, 1992). Because the stress reducing effort effects of noncompetitive, moderate exercise are well known (Shephard, 1997), we may indirectly infer that the relatively higher level of physical activity in CM as compared to SH, may explain the greater respiratory changes following CM.

During CM, subjects are given instructions to relax and maintain that mental state. It has already been shown that exercise plus cognitive strategy programs are more effective in promoting psychological benefits than are exercise programs lacking a structured cognitive component (Brown et al., 1995). Hence both cognitive and physiological factors may contribute to the greater reduction in oxygen and other respiratoy changes following CM as compared with supine rest.

In summary, the findings support the idea that CM, which combines "stimulating" and "calming" techniques, practiced with a background of relaxation and awareness may reduce physiological arousal better than SH can (which is calming). However, the present results require to be substantiated with added controls (e.g., studying the effect of no practice and of postures alone.

ACKNOWLEDGEMENTS
The authors gratefully acknowledge Anand Kumar who carried out the statistical analysis and Naveen Visweswaraiah for typing the manuscript.