Geomagnetic variations of partly interplanetary origin, with cyclic signatures in human affairs and pathology include the incidence of various diseases, regarding which this study of healthy subjects attempted to determine an underlying mechanism by worldwide archival and physiological monitoring, notably of heart rate variability (HRV). In the past half-century, the possible health and other hazards of natural, solar variability-driven temporal variations in the earth's magnetic field have become a controversial subject in view of the inconsistent results. Some well-documented claims of associations between geomagnetic storms and myocardial infarction or stroke have been rejected by a study based on more comprehensive data analyzed by rigorous methods - covering, however, only part of a solar cycle in only part of a hemisphere. It seems possible that inter-solar cycle and geographic variability, if not geographic differences, may account for discrepancies. Herein, we examine the start of a planetary study on any influence of geomagnetic disturbances that are most pronounced in the auroral oval, on human HRV. The magnetic field variations exhibit complex spectra and include the frequency band between 0.001-10 Hz, which is regarded as ultra-low frequency by physicists. Since the 'ultra-low-frequency' range, like other endpoints used in cardiology, refers to much higher frequencies than the about-yearly changes that are here shown to play a role in environmental-organismic interactions revealed by HRV, the current designations used in cardiology are all placed in quotation marks to indicate the need for possible revision. Whether or not this suggestion has an immediate response, we have pointed to a need for the development of instrumentation and software that renders the assessment of circadian, infradian and even infra-annual (truly low frequency) modulations routinely feasible. HRV was examined on the basis of nearly continuous 7-day records by ECG between December 10, 1998, and November 2, 2000, on 19 clinically healthy subjects, 21 to 54 years of age, in Alta, Norway. A geomagnetic record was obtained from the Auroral Observatory of the University of Tromsø. First, frequency-domain measures of HRV were compared for each person in 24-hour spans of high geomagnetic disturbance versus quiet conditions. Second, cross-spectra between geomagnetic activity and HRV measures were quantified via the squared coherence spectrum using 7-day time series. A 7.5% increase in the 24-hour average of heart rate, HR (P = 0.00020) and a decrease in HRV were documented on days of high geomagnetic disturbance. The decrease in HRV was validated statistically for the 'total frequency', 'TF' endpoint (18.6% decrease, P= 0.00009). The decrease in spectral power was found primarily in the 'circaminutan frequency', 'VLF' (21.9% decrease, P< 0.000001) in conjunction with the 'minutes-to-hours' component, ultra-low-frequency, 'ULF' (15.5% decrease, P= 0.00865) and circadecasecundan 'low frequency', 'LF' (14.2% decrease, P = 0.00187) regions of the spectrum. Power-law scaling of the power spectra did not show any statistically significant difference. It is noteworthy that most of the decrease in HRV, except for the circaminutan (VLF) component, was observed only in the season in which sunshine alternated with darkness (D/L), a finding suggesting a mechanism influenced by the alternation of light and darkness. The hypothesis of a light-dark-influenced magnetoreception was also supported by cross-spectral analysis. Group-averaged coherence at frequencies coincident with the geomagnetic Pc 6 pulsations (with periods ranging from 10 minutes to 5 hours) differed with a statistical significance (P < 0.000001) among the three natural lighting conditions, the association being weaker during UL or D/D than during D/L. By contrast, no statistically significant differences were found in terms of the circadian and circasemidian frequencies in relation to the alternation of sunshine with darkness or rather circannual rhythm stage. In conclusion, evidence is provided herein that an alteration of HRV is most apparent in the circaminutan ('VLF') region, which is clinically important, because a reduction in its power is a predictor of morbidity and mortality from cardiovascular disease. The circadecasecundan ('LF') component of HRV also decreased in association with geomagnetic disturbance, which may reflect an episodic alteration of arterial pressure related to changes in geomagnetic activity. Lastly, our study suggests the existence of a light-dark-influenced magnetoreception mechanism in humans involving mainly the Pc 6 band of the magnetic field.