How many bromeliads are there?

Using logistic models and a curated species list, this study projects that the total number of Bromeliaceae species ranges between 6,658 and 7,498, revealing that current descriptions account for only 49–55% of the family's diversity and highlighting Brazil, Mexico, and the Andes as key regions for future discoveries.

The Gist

Imagine the natural world as a massive, unfinished library. For centuries, librarians (scientists) have been rushing in to catalog the books (species) on the shelves. But here's the catch: we've only cataloged about half the books, and the shelves are still being built.

This paper is like a report card for the "Pineapple Family" (Bromeliads)—those cool, spiky plants you see in tropical forests and as houseplants. The authors, a team of botanists, asked three big questions:

1. Who has been doing the cataloging over the last 250 years?
2. How many more books are hiding in the stacks that we haven't found yet?
3. Where should we look next to find them?

Here is the story they tell, broken down into simple parts:

1. The Changing of the Guard (Who is doing the work?)

Think of the history of discovering these plants like a relay race where the baton is passed between different countries.

• The Early Runners (1700s–1900s): The race started with European scientists. They were the first to spot these plants, mostly because they were the ones with the ships and the scientific institutions at the time.
• The Middle Runners (1900s–2000s): The baton was passed to the United States. American scientists took the lead, funding big expeditions and writing massive books about these plants.
• The Current Runners (2000s–Present): Now, the baton is in the hands of local experts from Latin America, especially Brazil and Mexico. This is a huge shift! It means the people who actually live where these plants grow are finally the ones leading the discovery. It's like the locals finally taking over the tour guide job instead of the tourists.

2. The Great Mystery: How Many Are Left?

The authors used a mathematical crystal ball (called a "logistic model") to predict the future. They looked at the speed at which new plants are being found and asked, "If we keep going at this rate, how many total plants are there?"

• The Current Count: We know of about 3,700 bromeliad species.
• The Real Count: The model suggests there are likely between 6,600 and 7,500 species in total.
• The Shock: This means we have only found about 50% of the family. Roughly half of the bromeliads are still "dark diversity"—they exist, but science doesn't even know their names yet.

It's like walking into a room with 100 people, but you've only met 50 of them. The other 50 are hiding in the corners, and we need to go find them.

3. The "Small Fish" Problem (Where are they hiding?)

The paper found a very interesting pattern: The bigger the fish, the easier it is to catch.

• The Past: In the old days, scientists found the "big fish"—plants that grew everywhere, were easy to see, and had huge ranges. These were the first to be named.
• The Present: Now, we are finding the "small fish." These are tiny, rare plants that only grow on one specific mountain peak or a single rocky cliff. They are hard to find because they are rare and live in places humans can't easily reach.

The map shows that the next big wave of discoveries will happen in Brazil, Mexico, and the Andes mountains. Specifically, look for them in the "inselbergs" (those giant, lonely rock mountains that stick out of the jungle) and the high, misty peaks where it's hard to hike.

4. Why This Matters

Why should you care about a plant you've never seen?

• The Race Against Time: These "small fish" (rare plants) are in danger. Because they only live in tiny areas, if we cut down the forest or build a road there, they disappear forever.
• The Tragedy: The scary part is that we might lose these species to extinction before we even know they exist. It's like burning down a library before we've finished reading the books.
• The Good News: The fact that local scientists are now leading the charge is a great sign. When local experts are funded and supported, they find these hidden gems faster.

The Bottom Line

The world of bromeliads is much bigger and more mysterious than we thought. We are currently in the "golden age" of discovery, but it's a race against the clock. The next great discoveries won't come from big ships in Europe; they will come from local scientists climbing the steep, misty mountains of South America to find the tiny, rare plants hiding in the shadows.

In short: We've found half the pineapple family, but the other half is hiding in the most beautiful, hard-to-reach places, and we need to find them before they vanish.

Technical Summary

1. Problem Statement

The scientific understanding of global biodiversity remains incomplete, particularly in the Neotropics, where species discovery rates lag behind extinction rates driven by habitat loss. While global estimates of total species exist, specific, robust estimates for taxonomic groups like the Bromeliaceae (pineapple family) are lacking. This uncertainty hampers effective conservation planning. The authors address three core questions:

• What are the historical patterns of bromeliad species description and authorship?
• How many bromeliad species remain undiscovered or undescribed, and how are they distributed across subfamilies, genera, and geography?
• Is there a correlation between the timing of species description and geographic range size (i.e., are small-ranged species described later)?

2. Methodology

The study employed a combination of data curation, historical analysis, and statistical modeling:

• Data Curation: The authors constructed a comprehensive dataset by extracting data from the Encyclopaedia of Bromeliads (Gouda et al., continuously updated), supplemented by Plants of the World Online (POWO), the International Plant Names Index (IPNI), and protologues.
  • Filtering: The dataset was refined to include only accepted species names (3,696 species). Subspecies, varieties, hybrids, and "inedit" names were excluded. For species lacking type locality data, geographic origin was inferred from current distribution for single-country endemics; multi-country endemics were excluded.
  • Authorship Analysis: Nationalities and lifespans of authors were compiled to track the shift in taxonomic effort over time.
• Statistical Modeling (Species Accumulation):
  • The authors modeled species accumulation curves using a time-explicit logistic function (following Mora et al., 2011): $N(t) = K / (1 + e^{-r(t-t_0)})$.
  • Parameters: $N(t)$ is the cumulative number of described species, $r$ is the intrinsic growth rate, $t_0$ is the year of peak description, and $K$ is the asymptote representing the estimated total species richness.
  • Fitting: Models were fitted using the nlsLM function in R with parametric bootstrapping to generate confidence intervals.
  • Extrapolation: The models were extrapolated 150 years into the future to estimate the asymptote ($K$).
• Range Size Analysis: A linear regression was performed to test the correlation between the year of description and species range size for 2,726 species with available distribution data (from Zizka et al., 2020).

3. Key Contributions

• Comprehensive Historical Record: The study provides the first complete, curated timeline of bromeliad species descriptions, distinguishing between new species descriptions and nomenclatural recombinations.
• Quantitative Projections: It offers rigorous, model-based estimates of the "dark diversity" (undescribed species) for the entire family and its major subfamilies.
• Shift in Taxonomic Power: It documents the historical transition of taxonomic leadership from Europe (18th–mid-20th century) to the USA (mid-late 20th century) and finally to Latin America (21st century).
• Discovery Patterns: It confirms the "small-ranged species last" hypothesis, demonstrating that recently described species have significantly smaller geographic ranges than historically described ones.

4. Key Results

• Description Rates: The rate of species description has accelerated dramatically, rising from an average of 0.57 species/year (1753–1800) to 33.8 species/year (2001–2025).
• Authorship Shift:
  • Europe: Dominated descriptions until the mid-20th century.
  • USA: Dominated the second half of the 20th century (heavily influenced by Lyman B. Smith).
  • Latin America: Now leads in the 21st century, particularly researchers from Brazil (1,143 descriptions) and Mexico (279).
• Estimated Total Diversity:
  • The logistic models project the total number of bromeliad species to be between 6,658 and 7,498.
  • This implies that only 49% to 55% of the family's diversity has been described to date.
  • Subfamily Breakdown:
    • Tillandsioideae: 54%–34% described (Estimated total: 2,702–4,351).
    • Pitcairnioideae: 60%–49% described.
    • Bromelioideae & Hechtioideae: Models could not converge due to steep, non-plateauing discovery rates, suggesting these groups have vast undiscovered diversity.
• Geographic Hotspots:
  • Recent discovery is concentrated in Brazil (+50% new species in the last 25 years), Mexico, and the Andes (Bolivia to Mexico).
  • The Guiana Highlands showed a decline in new descriptions after 2000, potentially due to access issues.
• Range Size Correlation: There is a significant negative correlation between the year of description and range size ($R^2_{adj} = 0.18, p < 0.001$). Widespread species were described early; recent discoveries are predominantly local endemics with small ranges.

5. Significance and Implications

• Conservation Urgency: The finding that nearly half of bromeliad diversity remains undescribed, combined with the trend that new species are small-ranged endemics, suggests a high risk of extinction before discovery. These species are often restricted to threatened habitats like the Brazilian inselbergs and Andean cloud forests.
• Taxonomic Capacity: The study validates the importance of strengthening regional taxonomic capacity. The shift to Latin American leadership in the 21st century correlates with increased discovery rates, suggesting that local expertise is crucial for uncovering remaining diversity.
• Methodological Validation: The study reinforces the utility of logistic modeling for estimating species richness in data-rich but incomplete taxa, while acknowledging limitations when discovery curves have not yet plateaued (as seen in Bromelioideae).
• Future Directions: The results identify specific geographic regions (SE Brazil, Andes) and taxonomic groups (large genera like Dyckia, Vriesea, Tillandsia) as priority targets for future biodiversity prospecting and conservation efforts.