Daily cycles of light and dark due to the rotation of the earth around its axis have dictated how life has evolved on Earth. Organisms have thus developed the ability to predict these 24-h cycles by developing an endogenous circadian clock, which is entrained to external cues. In mammals, circadian rhythms are controlled and generated by the central or master circadian clock located in the suprachiasmatic nuclei (SCN) of the anterior hypothalamus in the brain [1]. This master clock consists of multiple single-cell circadian oscillators that are synchronized to 24 h by environmental factors, primarily dark/light and also temperature and food. The retina perceives the dark/light cycle information and transmits this signal to the SCN via the retinohypothalamic tract. The SCN then transmits these inputs to peripheral oscillators located outside the SCN. Such peripheral oscillators exist in nearly all peripheral tissues such as liver, adipose, pancreas, muscle and blood. Therefore, circadian rhythms have been described for virtually all physiological and biochemical processes, including behavioural ones [2]. The SCN may influence the cardiovascular system primarily through multisynaptic neural projections to the heart, adrenal cortex, kidneys and vasculature and resultant neural or endocrine effects, and secondarily by circadian influences on behaviours such as sleep and activity and feeding patterns [3]. Thus, the master SCN clock sends periodic signals to the peripheral oscillators in order to prevent dampening of circadian rhythms, but a complex reciprocal relationship seems also to exist. Indeed, altered behaviours, for example when sleeping or food intake is desynchronized from normal circadian patterns, seem to affect the peripheral oscillators or clocks in particular, resulting in metabolic processes that are decoupled from the primarily light-driven SCN [4,5]. In addition, the multiplicity of peripheral clocks in the human organism has been found to comprise networks that both control circadian processes and are tightly connected to metabolic and neural control networks, such that disturbance of one affects the other. Hence, misalignment of circadian rhythms in the SCN and peripheral tissues may constitute an important underling factor for a variety of metabolic and vascular disorders such as obesity, metabolic syndrome, diabetes and hypertension [4,6,7]. Adverse cardiometabolic endpoints including higher glucose, insulin, leptin, ghrelin, glucagon-like peptide-1, cortisol, blood pressure (BP) and catecholamines have been reported as results of misalignments invoked by food intake/irregular meal frequency (for comprehensive reviews see, for instance [1,2,4]). This new evidence illustrates how knowledge on chronobiology, that is the circadian influences on mechanisms involved in metabolic and cardiovascular disease, and their interaction with behavioural influences, may be important for optimizing preventive lifestyle measures or pharmacological therapy [8,9]. For instance, preventing adverse behavioural triggers at specific vulnerable circadian times [4] or prescribing ingestion of medicines at specific times in anticipation of vulnerable periods [9,10] may maximize their beneficial effects. Currently, such timed measures or chronotherapy constitute new avenues of research for the prevention and treatment of many diseases, including asthma, arthritis, cancer and seasonal affective disorder [11]. Chronotherapy has been recommended for the treatment of major cardiovascular risk factors such as hypertension [11] and dyslipidaemia [10], though typically in the context of pharmacotherapy. Timing of meals may constitute one such lifestyle therapeutical tool as well. In fact, some form of dietary chronotherapy is already implemented in metabolic diseases such as diabetes, wherein the timing of meals is particularly important, as patients need to coordinate insulin administration with calorie intake. Interestingly, in recognition of the chronobiological aspects of nutrition, Franz Halberg [12] called for research on the ‘when to eat’ question more than two decades ago. Currently, the beneficial chronobiological effects of food intake (e.g. meal time, frequency and regularity) are being subject of intensive research as a potential additional or alternative tool to the ‘calorie in-calorie out’ model of weight management and treatment of obesity and related metabolic sequelae [2,4]. Some evidence pointing towards the benefits of a regular eating pattern and higher meal frequency seems to exist, though there are still considerable contradictions and surely very limited evidence with regard to the value of the time of day of food intake [2]. Given the established links between obesity and hypertension [13], any such beneficial effects would expectedly extend to hypertension as well, but studies examining the specific effects of chrobiological aspects of food intake on BP are considerably scarce. This issue of the Journal of Hypertension features an interesting study [14] on the associations of time-of-day of energy intake with prevalence and 10-year incidence of hypertension. In a large cohort of middle-aged individuals from the 1946 British Birth Cohort, Almoosawi et al.[14,15] examined the diet diaries of cohort participants at the age of 43 by dividing energy intakes into eight time-slots [prebreakfast and breakfast (merged into one breakfast slot), mid-morning, lunch, mid-afternoon, dinner, late evening and extras], and the proportion of energy intake consumed at each time slot was calculated. After adjustments for confounders such as sex, social class, smoking status, region of residence, BMI and total daily energy intake, the authors observed that higher energy intake at breakfast was associated with lower prevalence of hypertension. In prospective analyses, higher energy intake at late evening was associated with a higher incidence of hypertension and steeper increases in SBP and DBP. All together, and albeit not free of limitations (discussed below), these findings suggest that when we eat and not just how much we eat may be important to prevent hypertension. There are some limitations that warrant discussion, in hopes they may be overcome in future studies. At this point, it may be useful to establish that chronobiology refers to time, and time can be analysed in several ways: the clock time (e.g. meal at 0800 h), frequency (e.g. number of meals per day) and regularity (e.g. meals everyday at the same time) [2]. For instance, the time slots of the meals may need to be better anchored on clock time, as different individuals may take the same meal at considerably different times (breakfast at 0600 h or at 1000 h, for instance, due to different professional schedules, including shift work). Most importantly, the extent to which the beneficial effects of higher energy intake at breakfast and the deleterious effects of higher energy intake at late evening on BP can be simply attributed to the amount of calories rather than specific nutrients or food items typically consumed during those time slots/meals needs to be clarified [1]. In an earlier report of the 43-year-olds of this cohort, the proportion of calories consumed at breakfast that derived from proteins, fat and carbohydrates were nearly 14, 30 and 56%, respectively, whereas for the late evening meal, the same proportions were nearly 12, 27 and 45%, respectively, plus ∼12% derived from alcohol [15]. The breakfast carbohydrates were associated with lower odds of the metabolic syndrome (MetS) at the age of 53, which was explained by their beneficial associations with the abdominal obesity and elevated triglycerides traits of the MetS, but not with elevated BP; nutrient proportions at late evening were not associated with the MetS or its traits, however [15]. After-dinner snacks tend to consist of high-calorie or salty foods such as chips or peanuts eaten while sitting in front of the television or computer; both salt intake and associated sedentary behaviour could have thus explained the association between late evening energy intake with incident hypertension. Also, the extent to which the different frequency of meals between individuals could have accounted for the results, for instance, because some could have followed the classical three meals per day scheme of breakfast, lunch, dinner, whereas others had more meals (albeit the adjustment for differences in total daily energy intake), is not clear. A randomized controlled trial of reduced meal frequency without calorie restriction in healthy, normal-weight, middle-aged adults showed significant increases in BP when individuals consumed one meal per day than three meals per day; however, the effects could in part have been due to circadian variation in BP (vide infra) because BP measurements were obtained in late afternoon in the one meal per day vs. early morning in the three meals per day groups [16]. It is also possible that the beneficial effects of higher energy intake at breakfast reflected a lower proportion of breakfast skippers among normotensive individuals. In this line, intake of breakfast cereal, in particular whole grain, has been linked to lower incidence of hypertension and this beneficial association increased with increasing servings per week [17]. A recent trial comparing the 12-h postprandial sympathetic and haemodynamic responses after high-protein and high-carbohydrate meals in a subgroup of individuals of the PROPRES study showed that decreases in BP were more pronounced on the high-carbohydrate than the high-protein group on the first day of dietary supplementation, though the groups did not differ in this regard after 4 weeks, when reductions in BP were no longer significant in both groups. Noteworthy, in this study, the 12-h postprandial responses covered the periods, each of 4-h, after breakfast (set at 1000 h), lunch (1400 h) and dinner (1800 h), and although the authors did not compare the postprandial responses specific to each of these meals, their data clearly showed that declines in BP at both day 1 and after 4 weeks of the intervention, and in both groups, were steeper in the hours after breakfast than after lunch or dinner [18]. Therefore, nutrient quality and also the frequency and regularity of meals [2,19] could have explained the findings reported in the study by Almoosawi et al.[14]. These aspects may also limit the extrapolation of its findings, because they differ considerably across countries and/or cultures [6]. Also, the extent to which BP readings were obtained at the same time of the day among all participants in this cohort study and, within these, at the same time of the day between baseline and follow-up examinations is not clear. Although any departures from similar times between and within individuals were likely random and, as such, have introduced nondifferential bias to the estimates reported, future studies may need to consider time of assessments more thoroughly in view of the circadian BP variability. BP follows a clear circadian rhythm, with a well documented day/night variation in humans. Especially since the introduction of ambulatory BP measurements (ABPMs), this circadian BP variation has become better understood, currently contributing gradually to better clinical decision-making [11]. Indeed, although the reference method for the diagnosis of hypertension still relies on (time-unspecified) readings of BP values as determined in the nurse's or physician's office, international guidelines for the diagnosis and management of hypertension recognize the additional clinical value provided by 24-h ABPM [20,21]. In individuals with normal BP and uncomplicated essential hypertension, BP is lower during the night sleep, raising sharply with morning awakening, reaching two daytime peaks, one during the first hours of diurnal activity and the other in the late afternoon/early evening with a small afternoon (postprandial) nadir in between [11]. Appreciation of 24-h BP patterns has led to the categorization of individuals on the basis of the diurnal/nocturnal mean BP ratio into nocturnal ‘dippers’ (typically, ratio of 10–20%) and ‘nondippers’ (if ratio <10%), the latter being shown to be at a considerably higher level of cardiovascular risk [11]. The normalization of the circadian BP pattern to a dipper profile has thus emerged as an additional therapeutical goal, and tailored pharmacological chronotherapy has been suggested to accomplish that goal [11]. ABPM has also proven useful for the diagnosis and monitoring of postprandial hypotension, which is common and an important cause of falls, syncope and stroke in geriatric populations [22]. Interestingly, BP decreases after breakfast rather than after lunch or dinner has recently been linked to higher cardiovascular mortality among the elderly [23]. A variety of external factors (e.g. ambient temperature/humidity, physical activity, alcohol or caffeine beverages, meal composition and sleep duration and routines) and internal (e.g. sex, ethnicity, autonomic nervous system, vasoactive hormones and haematological and renal variables) factors affect daily BP variability. Disentangling the relative contribution of internal and external factors to an individual's level of BP is thus difficult. Evidence supporting the role of an endogenous component of human circadian BP rhythms derived from the observation, in rodents, that lesions imposed to the SCN (the master circadian clock) abolished the circadian rhythms of BP and heart rate (HR), without affecting the 24-h sleep–wake and mobility cycles [24]. Two recent studies performed among humans are worth mention. Shea et al.[3] investigated the endogenous circadian rhythm of BP, using a forced desynchrony protocol, that is the gold standard to disentangle the endogenous influences of the circadian system from the behavioural or environmental ones. Autonomic, haemodynamic and haemostatic markers were assessed in normotensive adults across three complementary, multiday, in-laboratory protocols: a 38-h constant routine, a 196-h forced desynchrony protocol with seven sleep/wake cycles, and a 240-h forced desynchrony protocol with 12 sleep/wake cycles; behavioural and environmental influences were controlled and/or uniformly distributed across the circadian cycles and the core body temperature was used to determine circadian phases. Notoriously, similar circadian rhythms in SBP and DBP were found in all three protocols such that the respective SBP and DBP peaks occurred at a circadian phase corresponding to approximately 2100 h. SBP showed peak-to-trough amplitudes of 3–6 mmHg and DBP of 2–3 mm Hg. Moreover, the rhythm of BP was unrelated to phase differences between the circadian rhythms of BP and variables such as cardiac vagal tone, HR, cortisol, catecholamines or urine flow [3]. These findings illustrated the existence of a robust endogenous circadian rhythm in BP peaking at approximately 2100 h, suggesting that the endogenous BP rhythm is unlikely to underlie the well documented morning peak in adverse cardiovascular events. Endogenous circadian peaks in variables such as cortisol, platelet aggregability and rise in epinephrine were more coincidental with that morning peak. These authors also investigated the extent to which the reactivity to exercise of the same autonomic, haemodynamic and haemostatic markers was altered at different circadian phases [8]. They observed circadian modulation of the cardiovascular reactivity to exercise, with greatest vagal withdrawal and peaks in catecholamines during the morning, factors that could possibly explain why this period seems to be a vulnerable time of day for cardiovascular events. Therefore, ideally, the study of time-of-day of energy and nutrient intake should be linked to 24-h BP readings for a better coverage of the associations between circadian rhythms in diet and BP. In the same line, concomitant assessment of physical activity levels [25] and sleep duration [5], factors that could have confounded the results of the present study, would be desirable. Nevertheless, the studies of Almoosawi et al.[14,15] have the merit of reminding us that current dietary guidelines are not yet specific regarding timing issues and that eating regularly at the right time of the day may be an underappreciated lifestyle therapy for the prevention of metabolic and vascular diseases. Indeed, although consulting dietary guidelines for the prevention and/or treatment of hypertension, one will find reduced salt and increased potassium intake, moderate alcohol consumption and adherence to Dietary Approaches to Stop Hypertension (DASH) or Mediterranean-type dietary patterns (which emphasize fruits, vegetables and low-fat dairy products; include whole grains, poultry, fish and nuts; and reductions in fats, red meat, sweets and sugar-containing beverages) as typical recommendations [21,26,27]. No specific indications with regard to the timing of meals or the distribution of the calories or nutrients among these throughout a typical day are given. Most importantly, the studies by Almoosawi et al.[14,15] illustrate how time of meals could be incorporated as a variable into large epidemiological studies examining links between nutrition and disease. Clearly, more research is needed, to accept or reject, with proper evidence, the famous advice of the late Adele Davis (1904–1974), a very popular (but also controversial) American nutritionist: ‘Eat breakfast like a king, lunch like a prince, and dinner like a pauper’. ACKNOWLEDGEMENTS I.F. is supported by a senior postdoctoral research grant (#2006T050) from the Netherlands Heart Foundation. Conflicts of interest There are no conflicts of interest.