Medicinal plants contain pharmacologically active substances that are responsible for their therapeutic effects. It is well known, for instance, that the infusion of Salix alba bark is able to reduce headache and fever: this is due to its content of salicylic derivatives, that are metabolised in the body to salicylic acid, a substance that inhibits the inflammatory response blocking the cyclooxigenase. It is interesting to note that salicylic acid is the same metabolic product obtained after taking an Aspirin tablet, and this clearly shows that there are no fundamental differences in the mode of action of synthetic and natural drugs. Therefore, the same principles of pharmacodynamics and phatmacokinetics used for synthetic products have to be applied, as far as possible, to medicinal plants for their rational use. The first one is that also medicinal plants can cause side effects: Ginkgo can give haemorrhages, Hypericum can give phototoxicity, and many plants can give allergic responses and so on. The second fundamental principle is that for any plant a threshold dose exists for the therapeutic effect: if the administered amount of active principles does not reach the threshold value, no effect will be obtained. The most important consequence of these facts is that it is of high relevance to know the active substances responsible for the activity of a medicinal plant, and to produce herbal drugs and their preparations with definite content of these substances.

Although no basic differences exist between natural and synthetic drugs, in some cases medicinal plants present a higher pharmacological complexity than synthetic products. Medicinal plants are often characterised by the presence of a mixture of different active principles, each of them has a different pharmacological profile, and therefore they can fight an illness from different points of attack. A good example of this is given by Ginkgo: its leaves contain ginkgolides, that strongly inhibits the Platelet Activating factor and by this way reduce the viscosity of blood, bilobalide, that reduces the cellular damage from oxygen deprivation, and flavonoids, that have anti-oxidant and radical scavenging properties. All these quite different pharmacological properties play together and make Ginkgo particularly useful in the pathologies due to circulation deficiencies, such as arteriosclerosis, Alzheimer disease or venous insufficiency. In other cases the same class of compounds shows different pharmacological properties: Chamomile flavonoids are strong anti-inflammatory agents, more than indomethacin, but they has also smooth muscle relaxant properties, higher than papaverine, and antimicrobial activity. Therefore, Chamomile preparations are very useful in pathologies where inflammation is associated with cramps, such as gastritis, or with microbial disorders, such as stomatitis or vaginitis.

Furthermore, whereas synthetic drugs are usually designed to strongly inhibit (or stimulate) one single pathway involved in a pharmacological effect, natural compounds may exert a milder effect on a number different pathways. Drugs as indomethacin or acethylsalicilc acid reduce the inflammatory response by an almost complete block of the cyclooxigenase metabolization of arachidonic acid, but this has as a consequence also the well known side effects of these drugs. On the contrary, flavonoids inhibits moderately the cyclooxigenase but also, to some extent, the lipooxigenase; they block free radicals, they reduce the release of histamine, they stabilise neutrophils and so on. All these mild actions together gives to some flavonoids, such as apigenin, an anti-inflammatory potency higher than that of indomethacin, without the usual side effects of this drug.

Due to the complexity of the composition of medicinal plants, the experimental models used in the pharmacological research on natural drugs have to be a little different from those used for a single pure compound. The sophisticated in vitro methods of the modern pharmacology are usually not suitable in the field of medicinal plants and crude plant extracts or their fraction have to be tested by means of adapted pharmacological models.