If you imagine the droplet of milk hitting the surface, you'd expect it to flatten, but also to bounce too - not all of the energy's absorbed straight away.Some of the droplet will bounce straight upwards, and some will spread into a sheet at the edge of the drop as it flattens. Surface tension will prevent the sheet expanding ad infinitum, either the sheet will be pulled back into the drop, or break up. What you're seeing is surface tension breaking up a nice symmetrical sheet into droplets, like you might get when spraying water from a hosepipe. These small droplets then land, but are too small to splash, forming the pretty pattern. Of course, you need just the right combination of drop size and speed for this to happen, which will vary with the properties of the liquid.

As the drop hits the hard surface it spreads out (and slightly upwards) in a thin sheet. However the sheet thins as the radius increases, and eventually surface tension breaks it into droplets. Hence the ring of drops at a uniform radius. However the elasticity in the initial impact also bounces some of the liquid straight back upwards for a limited distance, and surface tension pulls this into a blob that eventually falls back. With soft surfaces, e.g. impact with another liquid, (a) the sheet forms off the crater rim, (b) the blob in the middle is powered upwards by the recoil of the depressed central surface, and (c) the depressed surface, and its recoil, set up a diminishing oscillation that propagates outwards and forms those characteristic ripples. By the way, the drop that bounces back up can sometimes bounce quite high, and be surprising pure: as I found out when carelessly pouring chlorine-rich liquor into a swimming pool and ruined some clothes!

I have found the website of a photographer who specialises in taking pictures of the type you saw in the textbook. He uses a variaty of liquids and lighting to produce some quite stunning images. Try looking at www.liquidsculpture.com/index.htm

The following answer was selected and edited by New Scientist staffThe pattern of droplets surrounding the larger drop of spilt milk is the result of corona splashing. The study of splashes has a long and distinguished history, dating back to the late 19th century. Arthur Worthington, who was one of the first people to systematically study drop impacts, made it his life's work, and his book A Study of Splashes (Longman, 1908) contains many fascinating photographs of splashes. Perhaps the best-known photographs of a corona splash are those captured by Harold Edgerton starting in 1936.When a drop hits a surface, it can either spread, splash or rebound. The outcome depends on a number of factors, including impact speed, size, density, viscosity and whether the surface is initially dry or damp. For dry surfaces, impact drop morphologies are also influenced by the roughness and "wetability" of the surface.On a typical dry, flat and relatively smooth bench, such as the type often found in kitchens, the moment the falling drop strikes the surface, portions of the liquid moving downwards are pushed outwards from beneath the collapsing drop and immediately begin to spread along the bench top. A narrow and highly curved neck therefore develops between the thin liquid layer and the largely uncollapsed spherical drop above it. Provided the impact speed is above a critical value and the surface is not too smooth, surface tension in the necked region gives rise to a force directed at an angle to the horizontal.In corona splashing, the upward component in the surface tension at the neck causes the leading edge of the radially spreading liquid film to bend upwards, giving rise to the formation of a "corona", or crown, of the type mentioned in the question. Excess pressure caused by the collapsing drop continues to push liquid into the walls of the crown as the impact unfolds, extending it outwards and upwards.As the crown rises and increases in size the rim slows and thickens, taking on a roughly toroidal shape. The underlying mechanism leading to the rim of the crown breaking up into many smaller droplets is not yet entirely understood. It is thought, however, that at the moment when liquid begins to feed into the walls of the crown, rough anomalies on the bench's surface initiate small ripples which rapidly grow and inevitably cause the rim of the crown to form cusps, which break up into thin jets directed upwards and away from the crown. Instabilities in the ejected jets from the rim of the crown cause a droplet to pinch off at each end, giving rise to the distinctive coronet (see second Photo in original post) that is responsible for the observed splash pattern.Shortly afterwards, the walls of the crown start to thicken as it slows before falling back onto the bench under gravity, together with what remains of the protruding jets, to form the larger central drop. The effects of surface tension are responsible for holding the collapsing coronet together during its final dénouement.Besides milk, corona splashing is found in many other liquids including water, various alcohols and some paints. However, the opaqueness of milk makes it easier to see the droplet ring resulting from a corona splash.Seán Stewart, The Petroleum Institute, Abu Dhabi, United Arab Emirates

Just to add an addendum on Sean's comments, at least some of the mechanism involved in the breakup of the corona into smaller drops appears to involve turbulent interactions with the surrounding atmosphere. Lei Xu, Wendy W. Zhang and Sidney R. Nagel at the James Franck Institute did an interesting experiement in which they varied the atmospheric pressure in which drops were splashing, and found that at a pressure of 17.2 kPa, a drop hitting a glass slide simply flattened out without any splash or corona at all. You can find their paper, which includes photographs of drops hitting slides at different pressures, at http://arxiv.org/ftp/physics/papers/0501/0501149.pdf.