Strictly  speaking, this simple circuit shouldn't work! How could anyone expect  an ordinary light dependent resistor photo cell to 'see' through a  fingertip in natural daylight and detect the change in blood flow as the  heart pulsates? The secret is a high gain circuit, based on a dual op  amp IC which can be either the low power LM358 or the JFET TL072. The  LDR is connected in series across the 9V battery supply via a 100kO  resistor (R1) and the minute signal caused by the blood pulsing under  the skin is fed to the non-inverting (+) input, pin 3, of IC1a via a  0.µF capacitor.
Pin 3 is biased by a high impedance voltage  divider consisting of two 3.3MO resistors. The feedback resistors to pin  2 set the gain to 11 times. The output of IC1a is fed via a 0.47µF  capacitor and 220kO resistor to IC1b. This is configured as an inverting  op amp with a gain of 45 so that the total circuit gain is about 500.  The output of IC1b is used to drive an analog meter which may be a  multimeter set to the 10V DC range or any panel meter in series with a  resistor to limit the current to less than its full-scale deflection.  The prototype used an old VU meter with a 47kO resistor fitted in  series.
Circuit diagram:

Heart Rate Monitor Circuit Diagram
Note that the unit was designed to use the  Dick Smith Electronics light dependent resistor (Z-4801). Other LDRs may  require a change in the value of resistor R1. A light source such as a  high brightness LED is not required. All that is needed is a reasonably  well-lit room, preferably natural daylight, to produce a healthy swing  of the needle. Only when the hands are very cold does it make it a  little more difficult to accurately count the pulses. To check your  heart rate, carefully position your thumb or finger over the LDR and  count the meter fluctuations for a period of 15 seconds. Then multiply  the result by four to obtain your pulse rate. The circuit can not be  used if you are walking or running, etc.
 

 
 
 
