The new microfluidic technique, described in an early view article in Analytical Chemistry can reduce the amount of time the experiments take to 2 hours compared with 12 hours using standard protocols. The new device, developed by Dr Friederich Simmel and Dr Tim Liedl of LMU Munich in Germany, also reduces the amount of sample needed to carry out the reaction which can be of critical importance when dealing with valuable DNA samples. The 'melting temperature' of a DNA duplex is defined as the temperature at which half of the double strands are separated into single strands and is a convenient measure of the stability of a duplex. Knowledge of the melting temperature of a DNA or RNA sequence is important for many applications, such as the polymerase chain reaction (PCR) DNA amplification process as the reaction is based upon denaturation and renaturation cycles. Melting temperature information is also useful in equalising the hybridisation strengths of the oligonucleotides probes of microarrays and can be used to estimate genetic differences between species as well as detecting single-nucleotide polymorphisms Stability measurements can be used to estimate the amount of guanine and cytosine in a genome (GC content) with higher melting temperatures being associated with higher GC content. This is because the pairing between the guanine and cytosine bases is generally stronger than the pairing between adenosine and thymine. The common experimental method involves slowly heating (about 0.5ºC per minute) a sample of buffer solution and DNA to temperatures around 90ºC and measuring the absorbance at a wavelength of 260nm, which corresponds to the in-plane transitions of the bases electrons. The new method uses a linear gradient of formaldehyde created in a microfluidic chip to 'melt' fluorescently labelled DNA double strands and generate melting curves within 2 hours. Formaldehyde lowers the melting temperature of DNA with any given concentration being able to be mapped onto a corresponding virtual temperature. While the technique was able to detect differences in the length of complementary sequences that differed by only one nucleotide as well as detecting single nucleotide mismatches, the researchers wrote: "further improvements of the technique are necessary to circumvent the time and cost-intensive labelling of the DNA strands." They hypothesised that this could be achieved by: "the use of the double-strand specific dye SYBR green I, the use of molecular beacons, or the application of UV absorption measurements along the gradient."