In last year's journal a circuit for a batteryless flash slave was produced from the BCRA Electronics Group Newsletter. The brief description printed there was wrong and the (anonymous) contributor to this journal expanded this into a paragraph of complete nonsense. For the record, and so users can develop the circuit further, here is a more accurate description.
The circuit relies on the fact that the design of most flashguns ensures that a reasonably high voltage is available at the terminals, albeit at a high impedance. This can be seen from the figures below which give the circuits for both bulb and (simplified) electronic flashguns. This terminal voltage is used to power the slave flash trigger unit. In the circuit of the slave flash (reproduced below) the bridge rectifier produces a positive voltage from either polarity of input. It is required because you cannot be sure of the polarity at the terminals. The capacitor charges up via the 2OMOhm resistor and is clamped to 6.8v by the zener diode. If the diode were absent the potential would rise to several hundred volts and the 2OMOhm resistor is present so as to not load the flash gun's trigger circuit. The high value resistor requires a low value capacitor to keep charging time reasonably quick and this in turn means that there is not much charge available for any sophisticated circuits but there is plenty to trigger the SCR when the phototransistor conducts. The SCR only provides a unidirectional current path, but the bridge rectifier again ensures that the trigger current can flow.
The author also claimed that the circuit would not work with a bulb flash gun. There is no reason why it should not but the component values need to be optimised. A further development would be to include filtering components so that the slave was only triggered by a sharp pulse from an electronic flash gun, rather than by a caver's light shining on it. This additional circuitry can be built discretely or by using very low current op amps such as the Texas Lin CMOS type.