American Scientist

To the Editors:

In the article "Breeding Better Buildings" by Rafal Kicinger and Tomasz Arciszewski (November-December), I was struck by a caption encouraging "modern-day architects" to learn from nature. Perhaps modern-day architects could also do well by looking back a generation or two to see what nature revealed to those builders.

I offer a bit of architectural history. The 1930s Johnson Wax Building in Racine, Wisconsin, owes perhaps its very existence to nature-inspired architecture. Frank Lloyd Wright designed a great multistory open space with its ceiling supported by tall cylindrical columns. Although seemingly tubular, the exterior of these columns is not load bearing but only a sheath around a cylindrical web-like skeleton. The inspiration for the internal construction of these columns was the skeleton of the giant saguaro cactus. This design is very strong, highly resistant to compression and even lighter in weight than a continuous tube. The fundamental repeat unit also bears a striking resemblance to the diagonal cross-braces discussed by Drs. Kicinger and Arciszewski, but nature had invented it first.

Jared L. Rifkin
Queens College
The City University of New York

Drs. Kicinger and Arciszewski respond:

The examples Dr. Rifkin provides illustrate very well how engineers and architects have previously used the inspiration from nature on several levels.

The first type of inspiration is visual, based on imitating nature's shapes and patterns in engineering designs. A second type of inspiration is conceptual,  based on understanding nature's principles and applying them in the design process. These two levels of inspiration have been used by engineers and architects for many years, and we are aware of this fact.

What we discuss in our article is another level of inspiration from nature at the computational level. Here, we not only need to understand nature's essential processes, but we also want to simulate it on our computers. Achieving this level of inspiration from nature has been possible only recently because of progress in such fields of science as computational biology and computer science. As we attempted to show in our article, with this computational inspiration from nature, we could also produce interesting patterns that we then used in our designs.

Thus, if our tool can produce similar bracing patterns as previously discovered by nature, we may have a new method for simulating various processes in nature at the computational level and exploring their potential for producing innovative designs.

Moreover, as these processes are simulated in silico, we may even change the details of the biological processes used to achieve certain engineering objectives.