The world of astronomy is entering a new era of large optical telescopes — giant pieces of equipment with primary mirrors measuring eight or more metres across. The light-gathering capacity of these enormous 'photon buckets' gives them inherent advantages: astronomers can observe fainter sources, and therefore address problems that were beyond the capacity of smaller telescopes. They can also observe brighter sources much more quickly, speeding up project execution.

Add to that the fine resolution made possible by combining a large mirror with adaptive-optics techniques, which correct for the distorting effects of atmospheric turbulence, and it is easy to see why the operators and users of small telescopes feel threatened. Money for astronomy comes from a limited pot, and with titanic telescopes consuming an ever-bigger share of it, many smaller observatories are coming under pressure to close (see page 12).

But although some older telescopes will have to go, it would be a mistake to cull the majority. For a start, many projects don't require giant mirrors, and can be efficiently conducted using a modest-sized telescope. And any study requiring repeated observations of the same object can only realistically be done using a small telescope. If the discoverers of extrasolar planets had had to compete for observing slots on one of the 10-metre Keck telescopes, for example, their high-risk endeavour would never have been granted enough time to bear fruit.

There is also a social factor: the shift towards fewer, larger telescopes inevitably places power in the hands of a privileged élite of astronomers. If diversity is strength, then, for the good of the astronomical community, it is important to retain a sufficient number of smaller telescopes to allow independent groups to do their own thing.

But given limited budgets, the users of small telescopes will have to look beyond a 'business as usual' approach. The small telescopes that survive into the era of the giants will be those that best demonstrate their efficiency and scientific productivity. That may mean that telescopes will have to concentrate on focused projects, rather than holding open competitions for observing time. It may also mean that ownership of these telescopes should pass from centralized organizations with relatively high overhead costs to leaner and fitter consortia of universities or research groups.

Finally, astronomers will need to change their attitudes towards the technical specialists who build instrumentation. If small telescopes are to work on focused projects, they will need to be fitted with highly specific instruments, optimized for the task in hand. Currently, those who build detectors and other instruments are sometimes treated as second-class citizens by certain of their colleagues. Their skills should be granted the respect already enjoyed by the builders of instruments in other scientific disciplines such as space science and high-energy physics.